The Philosophy of the Face and 20Th Century Literature and Art

Home , Characterology, Maximilian de Gaynesford, Otto Weininger, Real Face, Tone (literature), Vignette (literature)

University of Pennsylvania ScholarlyCommons

Publicly Accessible Penn Dissertations

2005 The hiP losophy of the Face and 20th Century Literature and Art Bernard J. Rhie University of Pennsylvania

Follow this and additional works at: http://repository.upenn.edu/edissertations Part of the Art and Design Commons, Comparative Literature Commons, English Language and Literature Commons, Esthetics Commons, Fine Arts Commons, Graphic Communications Commons, Neuroscience and Neurobiology Commons, Other History of Art, Architecture, and Archaeology Commons, Other Philosophy Commons, Philosophy of Mind Commons, Psycholinguistics and Neurolinguistics Commons, and the Theory and Criticism Commons

Recommended Citation Rhie, Bernard J., "The hiP losophy of the Face and 20th Century Literature and Art" (2005). Publicly Accessible Penn Dissertations. 1000. http://repository.upenn.edu/edissertations/1000

This paper is posted at ScholarlyCommons. http://repository.upenn.edu/edissertations/1000 For more information, please contact [email protected]. The hiP losophy of the Face and 20th Century Literature and Art

Abstract This dissertation explores the importance of the human face in modern literature, philosophy, and art. "Meaning is a physiognomy," wrote Wittgenstein--quite literally, if somewhat cryptically--in the Philosophical Investigations. My project takes this remark seriously and begins, in chapters one and two, by reading Wittgenstein's discussion of aspect-seeing alongside recent work in cognitive science and the philosophy of mind in order to explain how we perceive mentality in the appearance of a human face. I then trace the surprising ways in which our ability to understand facial expressions informs not only the way we understand language, but also other minds and the concept of personhood itself. Chapters three and four extend these findings into an analysis of visual portraiture, focusing on the paintings of Francis Bacon. Regarding the sense of injury often associated with Bacon's violently distorted likenesses, I ask why such "magical" feelings arise at all with respect to images of human faces. Reading Wittgenstein along with Gombrich and Wollheim, I find that the mind naturally responds to images of faces as expressive of mentality: we not only see faces in images but also to an extent see the images as persons. My final chapter looks into the ethics of physiognomy, asking what difference it makes whether we see the mind as a private substance or, as John Ashbery has suggested, a "visible core." This chapter reads two narratives about faces that dramatize the solipsistic consequences of a Cartesian commitment to mental privacy: that of the faceless woman in Rilke's Notebooks of Malte Laurids Brigge and the Magistrate's dreams about the tortured woman in Coetzee's Waiting for the Barbarians . I next consider Levinas' ethics of the face, an ambivalent critique of Husserlian phenomenology that tries, but ultimately fails, to escape this Cartesian predicament. In the end, the convergence of ethics and physiognomy may explain the face's importance to the modern imagination: perhaps, as Wittgenstein's writings suggest, faces grip us so because they call upon the same powers of pattern recognition that enable us to grasp the reality of other minds and moral values as well.

Degree Type Dissertation

Degree Name Doctor of Philosophy (PhD)

Graduate Group English

First Advisor Susan A. Stewart

Keywords physiognomy, portraiture, twentieth century

This dissertation is available at ScholarlyCommons: http://repository.upenn.edu/edissertations/1000 THE PHILOSOPHY OF THE FACE AND 20TH CENTURY LITERATURE AND ART

Bernard I. Rhie

A DISSERTATION

English

Presented to the Faculties of the University of Pennsylvania in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

2005

/I A'J / : / ' s Y /

Supervisor of Dissertation

Graduate Group Chair

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. COPYRIGHT

Bernard Jaeseong Rhie

2005

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS

Many people have helped me straggle with the questions that animate this

dissertation, so as I finish writing, I am delighted to have this opportunity to express my

gratitude for their encouragement and support. My warmest thanks go to the four

members of my dissertation committee. The deeply caring supervision of my adviser,

Susan Stewart, has been so essential to my development as a thinker that without her I

doubt this project would ever have been started, let alone finished. From early

conversations we had about ethics and phenomenology to detailed comments she later

wrote in response to drafts of every chapter, this dissertation and its author have

benefited immeasurably from her involvement. In so many ways she is an inspiration to

me, and I feel profoundly fortunate to have had a chance to work with her. I also feel

lucky to have benefited from the deep learning and theoretical sophistication of Jean-

Michel Rabate: his intimate familiarity with Continental traditions of philosophical

thought was an invaluable resource as I grappled with some of the fundamental

conceptual questions about language and subjectivity motivating this project. Bob

Perelman’s warm heart and good humor always helped keep things in perspective,

while his intimate knowledge of twentieth-century poetics enriched my own

understanding of the complex social dynamics informing the writing and reception of

poetry today. John Koethe, a professor of philosophy at the University of Wisconsin at

Milwaukee, influenced this project even before I met him: his excellent monograph on

Wittgenstein helped me formulate the reading of thePhilosophical Investigations that I

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. present in chapter two. 1 was therefore delighted when, after we met at a symposium on

contemporary poetics, he agreed to be a member of my dissertation committee. This

project has benefited immensely from his expertise in Wittgenstein and analytic

philosophy in general. Thinking of my four committee members, it’s difficult to

imagine a group of readers more attentive, responsive, and supportive than they have

been. That I knew they would be happy to look over whatever I sent them and quickly

write back with helpful criticisms and encouragement made the arduous process of

writing feel like an essentially intersubjective, rather than solipsistic, experience. For

that, and for everything else they have done for me, I thank them.

My good friend, Josh Kates, was not a member of my committee, but so

generously and carefully did he follow and respond to the argument of this dissertation

as it developed, that he played a role as crucial as anyone else to its successful

completion. In numerous phone calls, face-to-face conversations, and email

correspondences, I have benefited greatly from his profound knowledge of Husserl’s

writings and the phenomenological traditions it gave rise to. Jen Fleissner also read

through much of the dissertation as it came into being, and her friendship and

unflagging faith in me have nurtured me along the way. Ilan Sandler and Joshua

Margoiis read chapters one and two shortly after they were finished and provided

encouraging feedback that was much appreciated. Cindy Port, Julie Crawford, Liza

Yukins, Sari Gilman, and Alice Brittan have not read any of this manuscript, but their

sustaining friendship was essential nonetheless to its composition.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I would also like to thank a number of professors, in addition to those who

served on my dissertation committee, who have helped make my time at the University

of Pennsylvania so intellectually and personally rewarding. The many teachers I have

been fortunate enough to work with here have been just wonderful, as deeply humane as

they are learned. Special thanks go to Margreta de Grazia, whose imperturbable faith in

me helped see me through to the end. Other professors I would like especially to thank

are John Richetti, Chris Looby, Stuart Curran, Harold Bershady, Peter Stallybrass, Mark

Chiang, Jim English, Nancy Bentley, and Lisa New. For their support and for all that

they have generously taught me since I first arrived in Philadelphia, I am very thankful.

Family has been important too. For their patience and support over the many,

many years it has taken me to get to this point, I want to saythank you to my mother,

father, and sister. Mom and Dad, hooray: I’m finally done! And thank you also to my

father-in-law, Buddy Schutzman, and my mother-in-law, Jan Schutzman, for their

friendship and encouragement.

Finally, I would like to thank my wife, Julie Schutzman. In ways both small

and great, she has sustained me through both the thrilling and painful periods of the

writing process. Julie’s love and patience have made this work possible. But more

importantly, her presence in my life is what made this work feel meaningful to me in

the first place. It is, therefore, with all my love and gratitude that I dedicate this

dissertation to her.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

THE PHILOSOPHY OF THE FACE AND 20TH CENTURY LITERATURE AND ART

Bernard J. Rhie

Adviser: Susan Stewart

This dissertation explores the importance of the human face in modem literature,

philosophy, and art. “Meaning is a physiognomy,” wrote Wittgenstein—quite literally,

if somewhat cryptically—in the Philosophical Investigations. My project takes this

remark seriously and begins, in chapters one and two, by reading Wittgenstein’s

discussion of aspect-seeing alongside recent work in cognitive science and the

philosophy of mind in order to explain how we perceive mentality in the appearance of

a human face. I then trace the surprising ways in which our ability to understand facial

expressions informs not only the way we understand language, but also other minds and

the concept of personhood itself. Chapters three and four extend these findings into an

analysis of visual portraiture, focusing on the paintings of Francis Bacon. Regarding

the sense of injury often associated with Bacon’s violently distorted likenesses, I ask

why such “magical” feelings arise at all with respect to images of human faces.

Reading Wittgenstein along with Gombrich and Wollheim, I find that the mind

naturally responds to images of faces as expressive of mentality: we not only see faces

in images but also to an extent see the imagesas persons. My final chapter looks into

the ethics of physiognomy, asking what difference it makes whether we see the mind as

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a private substance or, as John Ashbery has suggested, a “visible core,” This chapter

reads two narratives about faces that dramatize the solipsistic consequences of a

Cartesian commitment to mental privacy: that of the faceless woman in Rilke’s

Notebooks of Matte Laurids Brigge and the Magistrate’s dreams about the tortured

woman in Coetzee’s Waiting for the Barbarians. I next consider Levinas’ ethics of the

face, an ambivalent critique of Husserlian phenomenology that tries, but ultimately

fails, to escape this Cartesian predicament. In the end, the convergence of ethics and

physiognomy may explain the face’s importance to the modem imagination: perhaps,

as Wittgenstein’s writings suggest, faces grip us so because they call upon the same

powers of pattern recognition that enable us to grasp the reality of other minds and

moral values as well.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table of Contents

Acknowledgements iii

Abstract vi

List of Illustrations ix

Preface 1

1. Face and Mind: Scientific and Philosophical Investigations 15

2. Wittgenstein’s Face of Meaning 61

3. Seeing As: Pictures, Perception and Persons 115

4. Bacon’s Sensational Portraiture 159

5. Face Value 209

Bibliography 283

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. List of Illustrations

Fig. 1 “Discrimination and imitation of facial expressions by neonates”p. 15

Fig. 2 A glass cube... an inverted open box... a wire frame of that shape p. 83

Fig. 3 Duck-Rabbit p. 86

Fig. 4 Picture-face, from Philosophical Investigations (p. 166) p. 142

Fig. 5 from Rodolphe Topffer, Essai du physiognomie p. 155

Fig. 6 Pablo Picasso, Self-Portrait, 1972 p. 161

Fig. 7 Pablo Picasso, Self-Portrait, 1907 p. 162

Fig. 8 Francis Bacon,Three Studies fo r Figures at the Base o f a p. 169 Crucifixion, 1944

Fig. 9 Francis Bacon,Three Studies o f Henrietta Moraes, 1969 p. 188

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Preface

Faces play a truly special role in human cognition. To begin with, no other

visual object communicates as many different types of social information. As Andrew

Young and Vicki Bruce put it:

We use them not only to recognise people we know, but also to infer moods and feelings, to regulate social interaction through eye contact and facial gestures, to assist in speech comprehension (e.g., when lip reading;... even people with normal hearing do this), to determine age and sex, and even to attribute characteristics on the basis of social stereotypes.1

One linguistic measure of the centrality of the human face to social life is its common

figurative employment as a synecdoche for the whole person. Consider such idiomatic

phrases as “face-to-face,” “say it to his face,” “set one’s face against,” “to lose face,”

“shut the door in his face” and “to show one’s face in public.” The referent of the word

“face” in these examples quite obviously extends far beyond the front of the human

head, rhetorically encompassing the whole person.

It is this everyday linguistic association between faces and persons that enables

literary and artistic representations of faces to serve so naturally as indices of

psychological states. Toni Morrison, for example, employs this logic in her novel,Song

of Solomon, at a moment when the main character, named Milkman, engages in a bit of

long-overdue self-scrutiny:

5 5 Young, A., Brace, V., “Perceptual Categories and the Computation of ‘Grandmother’” in Vicki Brace, ed., Face Recognition: a Special Issue o f the European Journal o f Cognitive Psychology 3 (1), pp. 5-49, p. 7. The parenthetical remarks on lip reading refer to a study by McGurk, H., MacDonald, J. (1976). “Hearing lips and seeing voices” inNature (264), pp. 746-8.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Milkman stood before his mirror and glanced, in the low light of the wall lamp, at his reflection. He was. as usual, unimpressed with what he saw. He had a fine enough face. Eyes women complimented him on, a firm jaw line, splendid teeth. Taken apart, it looked all right. Even better than all right. But it lacked coherence, a coming together of the features into a total self. It was all very tentative, the way he looked, like a man peeping around a comer of someplace he is not supposed to be, trying to make up his mind whether to go forward or to turn back.2

The apparent disjuncture of the features of Milkman’s face betrays a hesitant

personality that is developmentally incomplete, and thus manifestly out of joint. Here,

as elsewhere in Morrison’s novel, Milkman shows himself to be something of a modem

Hamlet, whose literary spirit haunts Morrison’s own meditation on ghosts and

intergenerational inheritance. The great themes of Shakespeare’s drama—the spectral

presence of a dead father, the troubled affective bonds between a mother and her son,

the inverse relationship between thought and action, and the psychological mood of

melancholy—all figure prominently in Morrison’s fable about African-American

masculinity. The importance of the face to its figurative logic as well is therefore

unsurprising, since the legibility of the face is naturally a question of great urgency in

Shakespeare’s drama of deceit and betrayal.

Morrison’s faith in the meaningfulness of the face’s appearance, however,

inverts the profound mistrust with which faces are consistently (and understandably)

regarded throughoutHamlet. Shakespeare’s play is, if anything, a skeptical meditation

on the incommensurability of surfaces and depths. In Claudius’ Denmark, one would

be a fool to trust appearances, especially those of a person’s face: after all, “one may

2 Toni Morrison,Song of Solomon (New York: Plume, 1977), p. 69-70.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. smile and smile and be a villain,” as Hamlet bitterly notes (1.5.115),3 Of course,

Claudius knows this as well, as he reveals in a telling aside that more than hints at the

deed that weighs heavily upon his soul: “The harlot’s cheek beautied with plast’ring art

/ Is not more ugly to the thing that helps it / Than is my deed to my most painted word. /

O heavy burden!” (3.1.59-62). Here the king’s public face is shown to be a kind of

mask, and we can see why in a world rendered so rotten by political theater, Hamlet

would perversely insist that the “fruitful river in the eye” and the “dejected havior of the

visage” should not be taken as reliable signs of his inner grief: “For they are actions that

a man might play” - “but the trappings and the suits of woe” - whereas he has “that

within which passes show” (1.2.83-88). Perhaps Shakespeare’s point is that politics is

but a debased form of theater and that therefore politics and the hermeneutic faith of

physiognomy are radically incompatible: after all, in another of Shakespeare’s great

political dramas we hear the ill-fated Duncan, who has just bestowed the title of the

Thane of Cawdor upon Macbeth, declare the ironically self-confirming sentiment that

“There’s no art / To find the mind’s construction in the face” (1.4.11-12).4

Unfortunately for the hapless king, Macbeth will soon conceal his own treachery with

“fairest show,” having learned that “False face must hide what the false heart doth

know” (1.7.82-83).

Claudius’ rhetorical play on conscience and cosmetics, and Hamlet’s obsession

with the difference between sincerity and theatricality, both suggest that faces can, and

often do, serve as deceitful masks. But, of course, masks need not be equated with

J William Shakespeare, Hamlet, Ed. Harold Jenkins (London: Routledge, 1982). 4 William Shakespeare, Macbeth, Eds. David Bevingtonet al. (New York: Bantam, 1980).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disguise, regarded, that is, asmerely theatrical, in the pejorative sense that is common in

the modem West. Shakespeare’s notion of deceit presupposes a doubling of the face

(captured nicely in the idiom “two-faced”); but what if there is no “inner” face beneath

which to search for that which “passes show”? Then, the face might be called a mask,

but a mask might also be understood as a kind of face: the distinction would lose its

hold on us. The ancient Greeks understood this well when they used the same word,

prosopon, to refer to both masks and faces. Eliot’s Prufrock, perhaps Hamlet’s canniest

modem reader, understood this too in his own neurotic way: “There will be time, there

will be time / To prepare a face to meet the faces that you meet” (11. 26-27).5 But even

when the conceptual line between face and mask becomes so utterly blurred, the face

nonetheless remains a privileged index of personhood, even if the person is now

regarded as apersona.

Twentieth-century investigations into the philosophy and science of mind tell us

a great deal about why the human face and personhood are so inextricably linked.

Much remarkable work has been done exploring the central role that faces play in

human life. Wittgenstein, especially, attributed great importance to the face in his later

philosophical reflections. In fact, the ways in which the workings of the whole body

make visible the life of the mind was an abiding concern of Wittgenstein’s later

philosophy; for as he declared in thePhilosophical Investigations, “the human body is

5 T.S. Eliot, “The Love Song of J. Alfred Prufrock” in The Waste Land and Other Poems, Ed. Frank Kermode (New York: Penguin, 1998), pp. 3-8.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the best picture of the human soul” (PI p, 152).° But it is clear that he regarded the face,

of all the parts of the body, as the most special. As we will see, Wittgenstein cared

deeply about understanding the nature of facial perception; however, he was interested

not only in how we see actual faces (cf., for instance, PI §§536-539), but how the way

we see faces informs numerous other aspects of human life as well, including the very

possibility of philosophical reflection itself. During his discussion of rule-following in

the Investigations, for example, we find the ancient art of face-reading invoked in order

to illustrate an important point about semantics: “Meaning is a physiognomy [Die

Bedeutung eine Physiognomie]” he concludes while musing upon the “character” of the

game of chess (PI §568). Physiognomy turns up again in the discussion of seeing

aspects and aspect-blindness in Part II (sec. xi), where he goes so far as to endow words

with faces: “The familiar physiognomy [Gesicht] of a word, the feeling that it has taken

up its meaning into itself, that it is an actual likeness of its meaning—there could be

human beings to whom all this was alien” (PI p. 186). The aspect-blind (those who

could not see a wordas its meaning) would regard words as purely arbitrary signs, the

connections between their letters and their meanings entirely conventional (as in

structuralist linguistics), just as those who look at faces un-physiognomically (like

Hamlet) would regard the relationship between a face’s appearance and a person’s

thoughts and emotions as contingent or even non-existent. For Wittgenstein, meaning is

to a word as mind is to a face; but he can express his understanding of how we

6 Ludwig Wittgenstein,Philosophical Investigations, 3rd ed., Trans. G.E.M. Anscombe (Oxford: Blackwell, 2001). References to this text will be given parenthetically with the abbreviation PI.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. experience the meaning of a word by likening a word to a faceGesicht ( ) only because

he already regards faces physiognomical!y: that is, as expressive of mind.

Just as a word, in our everyday use of language, is not taken as an arbitrary' sign

of a transcendent meaning, a face is likewise not an essentially “outer” sign to be

interpreted or deciphered for some contingently associated “inner” mental state. As he

puts it in a passage inZettel:

“We see emotion.”—We do not see facial contortions and make inferences from them (like a doctor framing a diagnosis) to joy, grief, boredom. We describe a face immediately as sad, radiant, bored, even when we are unable to give any other description of the features.—Grief, one would like to say, is personified in the face. (Z 220)7

Of course, someone may feign (or suppress) a facial expression to deceive us, just as

someone may lie with words (as with silence), yet the possibility of deceit is grounded

upon the prior and more fundamental ability of the face express:to “We say ‘The

expression in his voice wasgenuine’. If it was spurious we think as it were of another

one behind it.—This face he shews the world, inwardly he has another one.—But this

does not mean that when his expressiongenuine is he has two the same” (PI §606).

Mental states, as physiognomists have long insisted, are readily recognizable in the

faces of others:

Consciousness in another’s face. Look into someone else’s face, and see the consciousness in it, and a particularshade of consciousness. You see on it, in it, joy, indifference, interest, excitement, torpor and so on. The light in other people’s faces. Do you look into yourself in order to recognize the fury inhis face? It is there as clearly as in your own breast. (Z 220)

7 Ludwig Wittgenstein,Zettel, Eds. G.E.M. Anscombe and G.H. von Wright,, Trans. Anscorabe (Berkeley: UC Press, 1970). References to this text will be given parenthetically with the abbreviation Z.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wittgenstein’s later writings, that is, somewhat surprisingly ask us to take the art of

physiognomy quite seriously, certainly more seriously, I suspect, than most of us are

accustomed to doing.

For if physiognomists—from Aristotle to Montaigne, and Lavater to

Lombroso—have always tried to read the mind or soul of the individual exhibited

outwardly by the human face, our late modernity has tended to dismiss this desire as a

deluded, even pernicious, dream. Indeed, the art historian James Elkins declares the

practice of physiognomy a “dead issue” for contemporary art, and he even goes so far as

to characterize the twentieth-century in general as “antiphysiognomic”:

As a science, physiognomy has entirely vanished: not only do current art historians and artists not believe it but also they do not care about it, and contemporary artworks that concern the face avoid implying recoverable or specific psychological meaning. The twentieth-century is immoderate and skeptical.8

Of course, as Hamlet’s skeptical remarks about faces clearly suggest, doubts about even

its basic presuppositions have long dogged the practice of physiognomy and are

therefore far from being unique to the twentieth century. For instance, Hegel’s critique

of Lavater in the Phenomenology of Spirit was neither the first nor the last of its kind to

argue explicitly against the physiognomic assumption that “the reality of a man is his

face.”9 Nonetheless, our late modem disaffection with physiognomy is of an entirely

different, and I would argue higher, order. Hegel at least shared with Ms physiognomist

opponents a fundamental belief in the mental or spiritual expressiveness of the human

8 James Elkins, Pictures of the Body: Pain and Metamorphosis (Stanford: Stanford Univ. Press, 1999), p. 73. 9 G.W.F. Hegel, Phenomenology o f Spirit. Trans. A.V. Miller (Oxford: Clarendon Press, 1977), p. 193.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. face. As he remarked in hisLectures on the Philosophy of Fine Art, “Although the

expression of spirit must be diffused over the appearance of the entire body, it is most

concentrated in the face, whereas the other members can reflect the spirit only in their

position.”10 But during the twentieth-century, the once widely shared physiognomic

belief in the expressive meaningfulness of the face was cast into doubt. The explicit

affirmation of physiognomic views and ideas, especially in art and critical theory,

became questionable, even ideologically suspect. Of course, our everyday faith in

physiognomy remains untouched. As ever, people continue to interpret each other’s

faces on a daily basis, either tacitly or unconsciously. But a wide range of cultural

forms such as portraiture and the philosophy of art; poetry and poetic theory; even the

fields of semantics and ethics, all bear evidence of the problematic nature of the face in

modernity.

A number of factors have surely contributed to the attenuation or rupture of the

physiognomic link between the face and the meaningfulness of its expressions. Perhaps

many of us simply doubt whether the varieties of emotional experience considered

typical of modernity (such as anomie, angst, shock and trauma) can be expressedat all,

let alone by the features of the face. An absolute absence of expression, according to

this view, may very well characterize the modem face at its most “expressive,” calling

into doubt whether expression of any kind can convey what is most meaningful (or

meaningless) about our troubled modernity. Moreover, some particularly baleful

10 G.W.F. Hegel, Hegel’s Aesthetics: Lectures on Fine Art, Vol. 1. Trans. T.M. Knox (Oxford: Clarendon Press, 1975), p. 727.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. aspects of the twentieth-century are directly related to the ideological distortion and

misuse of physiognomy itself: for example, we might recall such repellent

physiognomic theories as Lombroso’s “criminal anthropology,” which correlated

craniofacial features to propensities for criminality, and Otto Weininger’sSex and

Character (1903), with its misogynistic and anti-Semitic theory of human morphology.

As recently as the 1920’s, Lombrosian measurements were admissible as prosecutorial

evidence in American courtrooms, while Weininger’s ideas were valued and

disseminated by Nazi propagandists.11 In the wake of such dark associations as these, it

is no wonder that physiognomy as a whole became suspect. And further contributing to

this theoretical distrust may be various forms of postmodern skepticism about

personhood itself: such skepticism, after all, entails doubt about the meaningfulness of

the face. In Paul de Man’s deconstructive notion that literary images of “defacement”

symbolize the figurative destiny of the psychological “subject” (itself regarded as

nothing but a linguistic trope), it is as if the truth of physiognomy were being affirmed

in an inverted and ironic manner, the fate of the face bound yet to the fate of the “inner”

person.

Whatever the reasons may be, the physiognomic face, understood as expressive

of a person’s character, thoughts and emotions, has now receded from sight into the

shadows cast by the doubts of a skeptical age. One measure of the widespread cultural

suspicion regarding physiognomy and the meaningfulness of the face is the frequency

with which modem artworks reduce the face to the plasticity of a mask or the

11 On Weininger’s use by the Nazis, cf. Ray Monk,Wittgenstein: The Duty o f Genius (New York: Penguin, 1990), p. 23.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. contingency of a sign, aesthetic tendencies exemplified by the silkscreens of Andy

Warhol and the photographs of Cindy Sherman. Early in the twentieth-century,

Guillaume Apollinaire’s musician from Saint-Merry “with no eyes no nose no ears” and

Giorgio de Chirico’s faceless mannequins prefigured but one limit of a generalized

occulting or eclipse of the face.12 And yet, despite the radical questioning of the

meaningfulness of the face, or indeed perhaps because of it, the meaning of the face as

such remains very much in question. Doubt, after all, is not the same as indifference.

That is, the face, even when treated with profound skepticism by modem thinkers and

artists, nonetheless remains a matter of great urgency, crucial to any understanding of

what it means to be human, including what it means to be able meanto anything at all.

In a foreword that J.M. Coetzee wrote for Cecile Pineda’s novelFace, a

harrowing tale of a man who becomes badly disfigured (in effect, faceless) after a

terrible accident, Coetzee suggestively remarks that one thing Pineda’s novel shows us

is that our culture has yet to give adequate thought to the meaning of the human face:

Helio Cara is a man who loses his face and leams what it is to live without a face in a society that is neither particularly cruel nor particularly kind, just has no philosophy of the face, has given no thought to the face, and therefore reacts to facelessness with bewilderment and anger.13

I doubt Coetzee himself believes that we have given no thought at all to the face:

indeed, as I will show in chapter five, his own novelWaiting for the Barbarians,

published four years before Pineda’s, is an important contribution to what might be

12 On Apollinaire’s faceless flute-player and his influence on surrealism and other modernist movements, see Willard Bohn,Apollinaire and the Faceless Man: The Creation and Evolution of a Modern Motif (London: Associated Univ. Presses, 1991). !3 J.M. Coetzee, “Foreward” in Cecile Pineda,Face, Rev. ed. (San Antonio: Wings Press, 2003), p. xi.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. called the moral philosophy of the face. Moreover, the rich insights about faces culled

from the writings of many others and gathered together in the pages of this dissertation

also belie the sweep of Coetzee’s remarks. Yet I would agree with the spirit of his

words. If the face has certainly been an object of reflection, nonetheless still much

more remains to be discovered, synthesized, and thought through before we can answer

such fundamental questions about the face as Coetzee poses here:

What is this thing, this structure of skin and bone and gristle and muscle, that we are condemned to carry around with us wherever we go? Where does it begin, where does it end? And why does everyone see it rather than seeing me? Or - turning the questions on their head: Who is this I that dares to think of itself as concealed behind its face, other than its face, so that its face is not it? (xi)

So with Coetzee’s remarks in mind, and with the intention of making a modest

contribution to some still incomplete philosophy of the face, this dissertation will

explore the important role representations of the human face play in twentieth-century

aesthetic experimentation and philosophical reflection.

My dissertation begins, in chapter one, with a survey of recent scientific studies

of the connection between face and mind. From the writings of investigators such as

Andrew Young and Martha Farah, I gather together research findings which show that

humans are biologically predisposed to treat faces and face-like visual patterns as

special perceptual objects. In addition, research on autism conducted by developmental

psychologists such as Uta Frith and Simon Baron-Cohen suggests that our innate

sensitivity to faces, in turn, plays an important role in the normal development of our

understanding of other minds as well as the acquisition of language. I conclude chapter

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. one by integrating these scientific findings about human psychology with philosophical

writings by Quine, Dennett and Davidson which offer a complementary, yet more

theoretically comprehensive, explanation of how we understand the behaviour,

linguistic utterances, and mental states of other persons.

Chapter two continues tracing the surprising ways in which our ability to

understand the expressions of the human face informs not only the way we understand

language, but also other minds and the very concept of personhood itself, attesting to

the deep truth of ancient correlations between faces, masks and persons. This chapter

supplements the findings of chapter one with a sustained reading of Wittgenstein’s

writings on aspect-seeing in order to more fully explain how we perceive mentality in

the appearance of a human face. Mental states are neither facts of the matter in the

features of the face (a face is only flesh, after all), nor subjective projections on the part

of the beholder, but rather perceptual phenomena—Wittgenstein calls them “aspects”—

that we naturally see by adopting a certainattitude towards the face’s appearance: “My

attitude towards him is an attitude towards a soul,” as theInvestigations puts it.

Wittgenstein’s discussion of aspect-seeing, like Dennett’s notion of the “intentional

stance” and Davidson’s theory of “radical interpretation,” shows that the human mind is

not a private Cartesian substance but what we “see” by virtue of a complex, though

utterly natural, process of intersubjective response and pattern recognition—and the

human face is the paradigmatic visual pattern in which such recognition of mind occurs.

Chapters three and four extend these findings into an analysis of visual

portraiture, focusing on the paintings of Francis Bacon. Beginning with Bacon’s

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. remarks during interviews with David Sylvester about the sense of injury often

associated with his physiognomic distortions, chapter three begins by asking why such

“magical” feelings arise at all with respect to visual representations of human faces.

Reading Wittgenstein with the work of art historians Ernst Gombrich and Richard

Wollheim, I find that the human mind naturally responds to images of facesas

expressive of mentality: we do not simply see facesin the images, but to an extent we

see, and thus respond to, the imagesas persons. Chapter four then goes on to argue that

the violent “sensations” Bacon’s paintings provoke is directly linked to this natural

cognitive confusion, explaining why Bacon’s art necessarily steered clear of total

abstraction: the affective force of his portraits depended upon the perceptual

recognizability of his faces as faces. Seen this way, Bacon’s portraiture comes into

focus as a sustained meditation on the nature of modem personhood in the wake of a

post-Darwinian acceptance of the contingency of life. I find, however, that a profound

ambivalence about personhood runs through Bacon’s portraiture: on the one hand, his

art affirms that personality is wholly visible, a cloud-like emanation that floats upon the

surface of a person’s face, while on the other hand, it implies that the self is a hidden

substance requiring an invasive and violent disclosure in order to be captured in paint.

In the end, though, this chapter argues that Bacon’s greatest obsession was with the

wonder of creation itself: that chance procedures in the studio might somehow replicate

and capture on canvas the total accident of personal existence.

My fifth and final chapter looks into the ethics of physiognomy, asking after the

moral consequences of whether we see the mind as a private substance or, as John

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ashbery has suggested, a “visible core.” This chapter reads two narratives about faces

that dramatize the solipsistic consequences of a Cartesian commitment to mental

privacy: that of the faceless woman in Rilke’s Notebooks ofMalte Laurids Brigge and

the Magistrate’s dreams about the tortured woman in Coetzee’sWaiting for the

Barbarians. Anticipating elements of Bacon’s portraiture, Rilke’s vignette is a gory

dramatization of the epistemological violence entailed by any conception of the mind as

essentially private. Coetzee’s novel too cautions us against such Cartesian imaginings,

suggesting that they are conceptually Isomorphic with the psychological views that

motivate Invasive interrogation practices used in the name of state power. I next

consider Levinas’ paradoxically iconoclastic phenomenology of the face, an ambivalent

critique of Husserlian phenomenology that aims, but ultimately fails, to escape this

Cartesian predicament. Levinas shows himself still to be in the grip of a fundamentally

Cartesian picture of mind, which explains why he would feel himself able to respect the

alterity of the other person’s face only by refusing to see it. Wittgenstein’s writings, on

the other hand, show how the expressive meaningfulness of the face can play a positive,

even essential, role in ethical life. I conclude by arguing that his theory of aspect-seeing

explains not only how we can understand the expressions of a human face but also how

we can perceive ethical truths, preserving thereality of moral claims while steering

clear of both moral subjectivism and fundamentalism.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1, Face and Mind: Scientific and Philosophical Investigations

Developmental psychologists have shown that we

enter the world predisposed to treat faces unlike all other

perceptual objects. Human neonates, for example, will track

a moving face further than any other pattern of comparable

complexity.14 In one study (by Gorenet al. [1975]), forty

infants with a median age of nine minutes were presented

with four different visual stimuli: a regular face pattern, two

scrambled faces, and a blank face. Each pattern was moved

back and forth in front of the newborns through a 180-

Fig. 1 degree arc, while their head and eye movements in response

to the stimulus were noted. Responsiveness to the regular face pattern was significantly

greater than to any of the other stimuli. As the researchers summarize the important

implications of their findings, which have since been replicated by others (Johnsonet

al. [1991]): “These results imply that organized visual perception is an unlearned

capacity of the human organism. The preference for the proper face stimulus by infants

54 These results were first reported in Goren, C.C., Sarty, M., and Wu, R.W.K. (1975). “Visual following and pattern discrimination of face-like stimuli by newborn infants”Pediatrics in 56, pp. 544-9. Goren et al.’s findings were replicated and expanded upon by Johnson, M. H., Dziurawiec, S., Ellis, H., Morton, J. (1991). “Newborns’ preferential tracking of face-like stimuli and its subsequent decline” inCognition 40 (no. 1-2): 1-19. Gorenet al. (1975) cited in Andrew Young,Face and Mind (Oxford: Oxford Univ. Press, 1998), pp. 71-2. Subsequent references to Face and Mind will be given parenthetically with the abbreviation FM. Johnsonet al. (1991) cited in Martha Farah,The Cognitive Neuroscience o f Vision (Oxford: Blackwell Publishers, 2000), pp. 116-7. Subsequent references toThe Cognitive Neuroscience o f Vision will be given parenthetically with the abbreviation CNV.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. who had not seen a real face prior to testing suggests that an unlearned or ‘evolved’

responsiveness to faces may be present in human neonates” (Gorenet aI, p. 544).

Moreover, in addition to showing an innate preferential attentiveness to face-like

patterns, humans also seem able to perceptually discriminate between different facial

expressions at a remarkably young age. One study (Fieldet al. [1982]) observed that

infants with an average age of but thirty-six hours could discriminate between at least

three different facial expressions (happy, sad and surprised).15 Repeated exposure to

one of these facial expressions modeled by a live adult would result in a gradual

diminishment of the infant’s visual attentiveness, while a presentation of a new

expression would result in a renewal of visual fixation, demonstrating that infant’s even

this young must be able to perceive differences among facial expressions. The

researchers also noted that the neonate subjects tended to imitate the facial expressions

with which they were presented (see fig. 1, from Fieldet al. [1982]), suggesting that

“there is an innate ability to compare the sensory information of a visually perceived

expression... with the proprioceptive feedback of the movements involved in matching

the expression” (Fieldet al., p. 181). These studies show that even at birth, humans

possess a special sensitivity to faces and their expressions.

As we grow older, faces retain their cognitive status as special perceptual

objects. Robert Yin, for example, demonstrated that our recognition of faces is

uniquely sensitive to spatial orientation. As Andrew Young describes Yin’s influential

work:

15 Field, T.M., Woodson, R„ Greenberg, R. and Cohen, D. (1982). “Discrimination and imitation of facial expressions by neonates,” Sciencein 281, pp. 179-81. Cited in Young,Face and Mind, pp. 72-3.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Yin (1969) began by comparing face recognition with recognition of other objects normally seen in one orientation (aeroplanes, houses, stick figures). Not surprisingly faces proved easier to recognize which, Yin realized, did not tell us anything very profound. Faces may be well remembered simply because of greater task familiarity or even because the particular set of faces used may have been especially easy to remember. For this reason Yin included a second condition in which the stimuli were inverted both at study and test. This condition produced a reversal in rank order, with faces now proving hardest to recognize.16

Yin hypothesized that this startling inversion effect indicated that faces are perceived in

a different manner than other classes of objects. Whereas most visual objects are

perceived by means of decomposition into discrete features, faces, Yin argued, are not

decomposed into parts and thus face recognition is disproportionately affected by

inversion.

Martha Farah and others have since provided confirmation of the hypothesized

link between the inversion effect and the special “holistic” nature of facial perception.

In one study, Farahet al. encouraged half of a group of human subjects to leam to

recognize a set of faces by familiarizing themselves with the various “parts” or

individual features of the faces. The other half of the subjects were presented with faces

presented in a normal format (that is, not broken up into parts). Remarkably, the

subjects who learned the faces in parts showed no inversion effect, while those who

learned the faces in a normal fashion did. Farah writes that “these results suggest that

what is special about face recognition, by virtue of which it is so sensitive to

16 Andrew Young,Face and Mind, p. 76. Citing Yin, R. K. (1969) “Looking at upside-down faces” in Journal ofExperimental Psychology (81), pp. 141-5.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. orientation, is that it involves representations with relatively little or no part

decomposition” (CNV 138).

Recent neuroanatomical studies in lower primates suggest that face perception is

special not only because of its holistic manner of visual processing, but also because it

is localized in different regions of the brain than are the systems for perceiving nonface

objects. A neuroscientific technique known as single cell recording, which measures

the neuronal activity of a given brain cell, has shown that there are individual cells

selectively responsive to the presence of face-like visual patterns. Face cells, as they

are called, “respond vigorously to faces yet show little response to nonface objects or

scrambled arrays of face parts” (CNV 119). In monkeys, face cells are particularly

concentrated in a single region of the brain (anterior superior temporal sulcus), where

they comprise up to 20 percent of all the cells in that area. While brain cells have been

found that are selectively sensitive to other types of visual objects, no other objects

except faces have so many cells devoted to their perception. Consistent with the

findings that facial perception is holistic in nature, “analyses of the aspects of facial

appearance that drive the cells suggest that configurational relations among multiple

facial features are critical” (CNV 119).17 However, it is not the case that a single cell is

devoted to a single face: while different cells respond differentially to different faces,

“each cell has a gradient of responsiveness to a range of faces, and any given face will

evoke activity at different levels over a subset of the face cell population” (CNV 121).

This suggests that face cells comprise a “system of distributed representation” in the

17 Citing Yamane, S., Kaji, S. and Kawano, K. (1988) “What facial features activate face neurons in the inferotemporai cortex of the monkey?” inExperimental Brain Research (73), pp. 209-214.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. brain, consistent with recent connection!st and neural network theories of cognition in

the philosophy of mind.

The specialness and specificity of facial perception is also indicated by the

nature of prosopagnosia, which is “an inability to recognize familiar faces following

cerebral injury” (FM 81) despite “intact intellectual functioning and even apparently

intact visual recognition of most other stimuli” (VA 69).18 Andrew Young explains that

people who suffer from this condition “are often unable to identify any familiar faces

including famous faces, friends, family, and their own faces when seen in a mirror...

They can distinguish faces as a category quite easily from other visual object... but have

no idea to which individual a specific face belongs” (FM 81). One study examined a

prosopagnosic (named W.J.) who happened to raise sheep. The researchers assembled

groups of photographs of three different classes of visual objects: human faces, sheep

faces of the same breed kept by W.J., and sheep faces of a different breed. They then

trained W J. along with a group of other, non-prosopagnosic, subjects to associate

names with each of the faces. The normal subjects performed better with the human

faces than with the sheep faces (even some who raised sheep themselves). W J., on the

other hand, performed poorly with the human faces and normally with the sheep faces.

Farah notes that the “data suggest that W.J.’s recognition impairment does not affect the

recognition of all groups of visually similar patterns, but is selective for human faces”

(CNV 123). This, and other types of related studies of normal and prosopagnosic

18 Martha J. Farah, Visual Agnosia: Disorders of Object Recognitionand What They Tell Us about Normal Vision (Cambridge, MA: MIT Press, 1990). References to this text will be given parenthetically with the abbreviation VA.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subjects, show that “prosopagnosia represents the selective loss of visual mechanisms

necessary for face recognition, and not necessary (or less necessary) for other types of

object recognition.... [TJhis implies that the face recognition system is separately

lesionable and hence anatomically distinct from the object recognition system” (CNV

127).

Findings such as these provide strong evidence that faces are in a perceptual and

cognitive category of their own. And indeed, it is not difficult to imagine evolutionary

reasons that such perceptual sensitivities would be selected for. The importance to an

infant’s survival of recognizing and responding to the faces of adult caretakers is

obvious. But, of course, our relationship to faces is far from entirely instinctual!

Surely, very few parts of the human body are as freighted with cultural associations as

is the face. One need only consider the types of headshots that are so common on the

covers of high-fashion magazines in order to appreciate the extent to which the

appearance of the face can be encoded with social meaning. But the evident extent to

which culture informs our responses to faces needs to be understood against the

background of scientific findings such as the ones we have already discussed; they

show clearly that the importance of the face to human forms of life has as much to do

with the kind of biological life we are as with what cultural forms we might socially

construct. The recent discoveries of cognitive neuroscience are important for

philosophically-minded reflection on the nature of human thought not only because of

what they can add to our knowledge about such issues as perception and imagination,

but also, and perhaps primarily, because of the constraints they put on our theoretical

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and philosophical speculations. If there are ways in which culture shapes our ideas

about faces - and certainly there aremany - that does not mean that the role of the face

in the lives of humans iswholly cultural, as has been maintained by some.

Gilles Deleuze and Felix Guattari, for example, argue inThousand A Plateaus

that the human face is ideologically determined, through and through.19 They see the

face as the repressive product of something they call an “abstract machine of faciality”

that, among other things, distributes to each member of a society “facial units” that are

structured by a binary semiotic logic: “a manor a woman, a rich person or a poor one,

an adult or a child, a leader or a subject, ‘an orx a y’” (TP 177). Sounding very much

like a postmodern parody of Lavater’s eighteenth-centuryEssays on Physiognomy, they

argue that the various “concrete individualized faces are produced and transformed on

the basis of these units, these combinations of units—like the face of a rich child in

which a military calling is already discernible, that West Point chin” (TP 177).20 In

other words, Deleuze and Guattari see the human face as a socialconstruct —“You

don’t so much have a face as slide into one” (TP 177). They therefore regard the face

as a thoroughly political issue. In accord with their analysis of the ideological

19 Gilles Deleuze and Felix Guattari,A Thousand Plateaus: Capitalism andSchizophrenia, Trans. Brian Massumi (Minneapolis: Univ. of Minnesota Press, 1987). References to this text will be given parenthetically with the abbreviation TP. 0 Johann Caspar Lavater,Essays on Physiognomy, 3 vols. (London: J. Murray, 1789-98). For example, here is a typical analysis of Lavater’s: “A portrait by Colla, which... we may affirm to be a great likeness. Nature, precision, harmony, exactness, are discoverable in every part. The flat, somewhat sinking forehead, agreeable to the whole, denotes an unpolished person, confined within a small circle of domestic economy. The strong eyebrows do not speak mental, but bodily power. Eyebrows are only significant of the former when they are unperplexed, equal, and well disposed. Nose, chin, neck, hair, all are characteristic of rude, narrow insensibility. Rustic sincerity is evident in the mouth” (p. 177). Cited in John Graham,Lavater’s Essays on Physiognomy: A Study in the Historyo f Ideas (Berne: Peter Lang, 1979), p. 46.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. constitution of “facial units,” they (logically enough) call for the “dismantling” of the

face (TP 188). Such a politics would amount to an overcoming of both the ideology of

subjectivity and the semiotics of the signifier that motivates the social constaiction of

human faces in the first place, for they explicitly position faces at the intersection of

language and subjectivity: the “faciality machine,” as they put it, is “the condition of

possibility” of both the “signifier” and the “subject” (TP 180). To dismantle one’s face

would therefore amount to breaking through both language and identity as we know it -

a heady possibility that Deleuze and Guattari find not only intelligible but deeply

attractive as well: “If the face is a politics, dismantling the face is also a politics

involving real becomings, an entire becoming-clandestine. Dismantling the face is the

same as breaking through the wall of the signifier and getting out of the black hole of

subjectivity” (TP 188). Deleuze and Guattari’s theory of “faciality,” however, is

predicated upon a thoroughly constructivist understanding of the role the face plays in

human societies, which therefore presupposes that that role might be fundamentally

altered. Recent work in cognitive neuroscience, however, enables us to see that such

views are entirely misguided: dreams of “dismantling” the face are but idle flights of

fancy. There can be no social constructivist understanding of the face.

• Life and Logos: Faces and the Development of Mind

In fact, not only do human sensitivities to the face clearly exist at birthprior - to

any kind of enculturation - we will see that such innate capacities actually enable and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. facilitate subsequent developmental steps towards full-blown socialization, If, as

Aristotle put it, humans are “rational animals”(zoon logon echon), our sensitivity to

faces can be understood to straddle or bridge the conceptual and developmental chasm

that separates the animal from the rational sides of our being. Aristotle’s apt

description of human beings can, of course, also be translated as speech-possessing life:

logos is at once reason and speech, and along with Aristotle countless philosophers have

claimed that the possession of language is a necessary condition of rationality. This is

Donald Davidson’s view, for example, and in an essay entitled “Rational Animals” he

explains why he thinks that rationality depends on the possession of language.21 To

begin with, he sees rationality as consisting in “having propositional attitudes such as

belief, desire, intention, and shame”: “to be a rational animal is just to have

propositional attitudes, no matter how confused, contradictory, absurd, unjustified, or

erroneous those attitudes may be” (SIO 95). A substantial degree of actual irrationality

in a given human is fully consistent with Davidson’s fundamental claim, for “the

possibility of irrationality depends on a large degree of rationality. Irrationality is not

mere lack of reason but a disease or perturbation of reason” (SIO 99). This is what

distinguishes the rationality of a normal adult from the mere animal existence of a snail

or an infant one week old:

Neither an infant 1 week old nor a snail is a rational creature. If the infant survives long enough, he or she will probably become rational, while this is not true of the snail. If we like, we may say of the infant from the start that he is a rational creature because

Donald Davidson, “Rational Animals”Subjective, in Intersubjective, Objective (Oxford: OUP, 2001), pp. 95-105. References to Subjective, Intersubjective, Objective will be given parenthetically with the abbreviation SIO.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. he will probably become rational if he survives, or because he belongs to a species with this capacity. Whichever way we talk, there remains the difference, with respect to rationality, between the infant and the snail on the one hand, and the normal adult person on the other. (SIO 95)

Davidson goes on to note that propositional attitudes only come “as a matched set”: “to

have one is to have a large complement. One belief requires many beliefs, and beliefs

demand other basic attitudes such as intentions, desires, and, if I am right, the gift of

tongues.... [T]he intrinsically holistic character of the propositional attitudes makes the

distinction between having any and having none dramatic” (SIO 96). But the key

element in Davidson’s argument is that the basic propositional attitude of belief itself

requires one to possess a conceptual distinction between “what is believed and what is

the case.” Interestingly, this makes the ability to experiencesurprise criterial of

rationality: “surprise about some things is a necessary and sufficient condition of

thought in general.” Davidson’s point is that the ability to feel surprise about one’s

beliefs (to find that a belief one holds is false) indicates that one possesses “the idea of

an objective reality which is independent of [one’s] belief’ (SIO 104). For Davidson,

this does notprove that linguistic communication is necessary for rationality (after all,

nothing could), but he also has no “idea how else one could arrive at the concept of an

objective truth” (105). He therefore concludes that “rationality is a social trait. Only

communicators have it” (105).

But how do we to come to possess the language that enables us to have a

concept of an objective truth in the first place? That is, how do we enter the space of

reasons, the interdependent web of propositional attitudes that constitutes rationality?

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In an essay on “The Emergence of Thought,” Davidson discusses this question, which

he acknowledges is a profoundly difficult one.22 For we are asking what happens to a

human infant between the time it is a non-rational animal and the time at which it

possesses the full-blown rationality of an intentional agent who possesses beliefs and

other related propositional attitudes. Because of the holism of the mental, as Davidson

terms the interdependency of the various aspects of mentality, it is difficult to describe

the intermediate steps that link the two sides of the Aristotelian phrase “rational

animal”:

We have many vocabularies for describing nature when we regard it as mindless, and we have a mentalistic vocabulary for describing thought and intentional action; what we lack is a way of describing what is in between. This is particularly evident when we speak of the “intentions” and “desires” of simple animals. We have no better way to explain what they do. It is not that we have a clear idea what sort of language we could use to describe half-formed minds; there may be a very deep conceptual difficulty or impossibility involved. (SIO 128)

For Davidson, meaning, like mentality, is holistic in nature: “words, like thoughts, have

a familiar meaning, a propositional content, only if they occur in a rich context, for such

a context is required to give the words or thoughts a location and a meaningful

function” (SIO 127). Therefore, if we happened to come upon a mouse who had the

right kind of vocal cords and trained it to say the word “cheese,” that “word would not

have a meaning when uttered by the mouse, nor would the mouse understand what it

‘said’” (127). The early verbal sounds emitted by an infant would be of the same

nature: “Infants utter words in this way; if they did not, they would never come to have

22 Donald Davidson, “The Emergence of Thought” in SIO, pp. 123-134.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a language. But if you want to describe what is going on in the head of the child when

it has a few words which it utters in appropriate situations, you will fail for lack of the

right sort of words of your own” (SIO 127-8).

While he acknowledges the profound difficulties involved, Davidson

nonetheless goes on to suggest one developmental step, which he calls “triangulation,”

that he sees as a “necessary condition for thought and language, a condition that can

exist independent of thought, and can therefore precede it”:

The basic situation is one that involves two or more creatures simultaneously in interaction with each other and with the world they share; it is what I call triangulation. It is the result of a threefold interaction, an interaction which is twofold from the point of view of each of the two agents: each is interacting simultaneously with the world and with the other agent. To put this in a slightly different way, each creature learns to correlate the reactions of other creatures with changes or objects in the world to which it also reacts.... [T]he triangleI have indicated is essential to the existence, and hence to the emergence, of thought. For without the triangle, there are two aspects of thought for which we cannot account. These two aspects are the objectivity of thought and the empirical content of thoughts about the external world. (SIO 128-9)

Tri angulation is a situation of shared attention and response: two creatures attending

and responding to the same objects or events in the external world while, at the same

time, being aware that they are doing so. If we imagine the reactions of one of the

agents as taking a linguistic form (“Here’s your bottle”), then we have a picture of the

very common way in which infants are taught numerous basic words, such as milk,

mommy, doggie, or ouch.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Those familiar with the work of Quine will recognize the basic outlines of this

scenario from the “radical translation” thought experimentWord in and Object, which

had a profound influence on the development of Davidson’s ideas."9^ There, Quine

imagines a situation in which a linguist attempts to translate a foreign language for

which no translation manual exists:“radical translation, i.e., translation of the language

of a hitherto untouched people” (WO 28). Quine notes that the kinds of sentences most

likely to be translated first under such conditions would refer to “present events that are

conspicuous to the linguist and his informant,” to events, that is, that can be readily

triangulated in Davidson’s sense: “A rabbit scurries by, the native says ‘Gavagai’, and

the linguist notes down the sentence ‘Rabbit’ (or ‘Lo, a rabbit’) as tentative translation,

subject to testing in further cases” (WO 29). In order to be sure that the one-word

observation sentence “Gavagai” means the same thing as the English “Rabbit,” the

linguist must repeatedly test the correlation, stating “Gavagai” himself in the presence

of a rabbit while checking to see if this utterance consistently prompts the native’s

assent. If the correlation is confirmed frequently enough, the linguist can conclude that

the meanings of the two observation sentences are the same. Unlike Davidson, who did

not follow him on this score, Quine views the meaning of observation sentences as the

disposition of speakers to utter certain sounds in response to certain patterns of sensory

stimulation: thus Quine calls the meanings of observation sentences “stimulus

meanings” (Davidson, on the other hand, came to view the distal object itself, rather

than the proximal pattern of sensory stimulation, as the third point of the triangulated

23 Willard Van Orman Quine,Word and Object (Cambridge, MA: MIT Press, 1960). References to this text will be given parenthetically with the abbreviation WO.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. relationship24). In Word and Object , therefore, Quine states that the sameness of the

meaning of the two sentences (“Gavagai” and “Rabbit”) is attributable to the sameness

of sensory stimulation for both speakers (WO 43). One point of the thought experiment

was to show that it was possible to explain linguistic meaning and translation simply by

reference to the relationship between behavior and patterns of stimulation; in other

words, Quine saw no need to posit the existence of non-physical (mental or intentional)

objects in order to explain effective translation. As Quine later put it:

The point of my thought experiment in radical translations was philosophical: a critique of the uncritical notion of meanings and, therewith, of introspective semantics. I was concerned to expose its empirical limits. A sentence has a meaning, people thought, and another sentence is its translation if it has the same meaning. This, we see, will not do.25

Quine’s point, of course, went beyond the situation of translation and was intended to

account for linguistic meaning in general: “There is nothing in linguistic meaning

beyond what is to be gleaned from overt behaviour in observable circumstances,” as he

later put it (PoT 37-8). Quine, however, is no crude behaviorist: he does not deny the

existence of mental states that are inner and not publicly observable. Rather, he merely

notes that “as long as our command of our language fits all external checkpoints, where

our utterance or our reaction to someone’s utterance can be appraised in the light of

some shared situation, so long all is well. Our mental life between checkpoints is

indifferent to our rating as a master of the language” (PoT 37-8). Because the meanings

24 On the differences between Quine and Davidson on this matter, see Quine’sThe Pursuit o f Truth (Cambridge, MA: Harvard Univ. Press, 1990), pp. 41-2. Subsequent references to this text will be given parenthetically with the abbreviation PoT. 23 Quine, W.V.G. “Indeterminacy of translation again” (1987), p. 9.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of sentences are not intentional objects, it follows that there is no fact of the matter

whether a given sentence in one language correctly translates a given sentence in

another. If a translation works, it works: the pragmatic fact that it works is the only

“fact” that might settle the matter. It is conceivable, therefore, that two radical

translators might produce perfectly adequate, but incompatible, translation manuals for

a single native language. This possibility is what Quine called the thesis of the

indeterminacy of radical translation.

However, the positing of an intersubjective likeness of stimulation for the two

speakers in the radical translation thought-experiment posed conceptual problems that

Quine eventually came to view as intractable. As he later put it:

an event of stimulation... is the activation of some subset of the subject’s sensory receptors. Since the linguist and his informant share no receptors, how can they be said to share a stimulation? We might say rather that they undergo similar stimulation, but that would assume still an approximate homology of nerve endings from one individual to another. Surely such anatomical minutiae ought not to matter here. (PoT 40)

In The Pursuit of Truth, Quine therefore decided to discard the notion of likeness of

stimulus meaning first presented inWord andObject: “The view that I have come to,

regarding intersubjective likeness of stimulation, is rather that we can simply do without

it. The observation sentence ‘Rabbit’ has its stimulus meaning for the linguist and

‘Gavagai’ has its stimulus meaning for the native” (PoT 42). The linguist observes the

native informant consistently assenting to “Gavagai” when he imagines that he himself,

in the same position, would do so as well. The linguist’s grasping of the likeness of

meaning is therefore the result of an “uncanny” kind of interpersonal imaginative

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. projection, which Quine termsempathy : “We all have an uncanny knack for

empathizing another’s perceptual situation, however ignorant of the physiological or

optical mechanism of his perception” (PoT 43). Our knack for empathy obviates any

need to make reference to “anatomical minutiae” of any sort. Quine then goes on to

offer a striking analogy in order to illustrate what he means by empathy: “The knack is

comparable, almost, to our ability to recognize faces while unable to sketch or describe

them” (PoT 43). It is by means of such empathy, Quine explains, that the linguist can

come to feel that the two words mean the same thing. It follows that empathic response

must be an essential component ofall language acquisition:

Empathy dominates the learning of language, both by child and by field linguist. In the child’s case it is the parent’s empathy. The parent assesses the appropriateness of the child’s observation sentence by noting the child’s orientation and how the scene would look from there. In the field linguist’s case it is empathy on his own part when he makes his first conjecture about “Gavagai” on the strength of the native’s utterance and orientation, and again when he queries “Gavagai” for the native’s assent in a promising subsequent situation. (PoT 42)

It is unclear why Quine would feel compelled to specify which party (child or parent,

linguist or native) would need to exercise empathy for effective language-learning to

occur, for it is obvious that in each case the empathy of both parties is necessary.

Imagine the linguist attempting to prompt the assent of the native informant to an

utterance of “Gavagai” if the native happened to be indifferent to the state of the

linguist’s mind. And what good would the empathy of the parent be without the

simultaneous and mutual attentiveness of the child? Any parent knows how pointless it

is to point to an object and name it if the child is not “tuned in” at the same moment.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. And certainly young children are familiar with the frustration of questioning an adult

about the name of an object when that adult’s attention is directed elsewhere.

Davidson’s model of tri angulation, therefore, while based on Quine’s discussion

of radical translation, better captures the mutuality of both agents necessary for

linguistic training to occur. But Quine’s point about the similarities between empathy

and facial perception provides a fascinating clue as to how the experience of mutuality

itself might arise. As it so happens, recent findings in developmental psychology

(especially concerning the condition of autism) support Quine’s intuitive association

between language learning, empathy and facial perception. If we consider it, the way

we can empathetically grasp a person’s state of mind in the appearance of his or her

face is a remarkable ability. As everyday as the experience is, how is it that we so

readily know when another person is attending to the same event or object that we are?

Davidson’s presentation of the idea of tri angulation presupposes the presence of this

social and cognitive capacity, and yet if we consider how remarkable that ability truly

is, it is clear that it is hardly something to be taken for granted. Indeed, research by Uta

Frith, Alan Leslie, Simon Baron-Cohen and their colleagues into the disorder of autism

has shown that it is this very ability to experience what they call “joint attention” that is

severely impaired or even absent in autistic children.26

26 See Uta Frith, Autism: Explaining the Enigma, 2nd ed. (Oxford: Blackwell, 2003); Alan Leslie, “Pretense and representation: The origins of ‘theory of mind’”Psychological in Review (94), pp. 412-26; and Simon Baron-Cohen,Mindblindness: An Essay on Autism and Theory of Mind (Cambridge, MA: MIT Press, 1995), pp. 64-96. Subsequent references to Frith’s text will be given parenthetically with the abbreviation AEE; subsequent reference to Baron-Cohen’s text will be given parenthetically with the abbreviation MB.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. • Seeing Eye to Eye: Autism, Mindblmdness and the Intentional Stance

The explanation of autism developed by Frith, Leslie and Baron-Cohen, widely

acknowledged to be the best one currently available, sees it as an impairment in the

normal human ability to “mind-read”: that is, to regard other humans as intentional

agents interpretable according to mentalistic descriptions. Children with autism, Baron-

Cohen explains, fail to develop this basic ability and therefore suffer from what he and

his colleagues calls “mindblmdness”: that is, they do not see others, to use Dennett’s

phrase, as “intentional systems.” In fact, Baron-Cohen’s explanation of autism adopts

wholesale Dennett’s notion of the “intentional stance,” and the great clinical success he

has had using Dennett’s theory provides some of the most compelling empirical

evidence in favor of its explanatory power. Baron-Cohen glosses the intentional stance

as our ability to attribute to others “the full set of intentional states (beliefs, desires,

thoughts, intentions, hopes, memories, fears, promises, etc)” (MB 21). Dennett himself

describes the intentional stance this way: “the tactic of interpreting an entity by

adopting the presupposition that it is an approximation of the ideal of an optimally

designed (i.e. rational) self-regarding agent” (CPM 238).27 Dennett’s formulation may

recall the sort of rational agent that is familiar from decision theory, which, as Davidson

points out, is “often derided as a false description of how people actually act” (SIO

126). However, Davidson believes such criticisms miss the mark:

2/ Ed. Samuel Guttenplan,A Companion to the Philosophy o f Mind (Oxford: Blackwell, 1994). References to this text will be given parenthetically with the abbreviation CPM.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ... the theory answers to our intuitions about how actual decisions are made; in effect it simply spells out our commonsense apparatus for explaining intentional action. For we all, whether we think about it or not, make our decisions in terms of how we weigh the values of various possible outcomes of our actions, and how likely we think one or another course of action is to attain those values. We understand why someone acts as he does only by supposing that he or she values to various degrees the possible results of action, and how probable he or she thinks a given action is to produce one or another result. So although not capable of precise predictions of actual choices, decision theory nevertheless corresponds to our intuitions about how actual decisions are made, and so is part of our commonsense apparatus for explaining intentional behavior. (SIO 126)

Consistent with Davidson’s talk of a “commonsense apparatus,” Dennett calls the

“intentional stance” a version of “folk psychology” (a phrase that Dennett himself

coined and which has become a major subject of debate in the analytic philosophy of

mind).28 While we may become philosophically reflective about what is entailed in

taking up the intentional stance, Dennett sees such theoretical explanations as

Davidson’s and his own as nothing but the explicit elaboration of what all normal

humans already do implicitly without thinking twice about it.

Dennett, along with Quine and Davidson, is a non-reductive physicalist (a

position distinguished, on the one hand, from mind-body dualism, and on the other,

from eliminative materialism, which sees “folk psychology” as illusory and dispensable,

28 Barbara Von Eckardt provides the following bare-bones preliminary description of what the phrase “folk psychology” means in the current philosophy of mind literature: “Human beings are social creatures. And they are reflective creatures. As such, they continually engage in a host of cognitive practices that help them get along in their social world. In particular, they attempt to understand, explain, and predict their own and others’ psychological states and overt behaviour; and they do so by making use of an array of ordinary psychological notions concerning various internal mental states, both occurent and dispositional. Let us then consider [folk psychology] to consist,at a minimum, of (a) a set of attributive, explanatory, and predictive practices, and (b) a set of notions or concepts used in those practices” (CPM 300).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ultimately to be superceded by scientific explanations2*). While there are important

differences among their individual versions of non-reductive physicalism, they are each

fully committed to the idea that there is nothing in the universe but entities of the Mud

posited in (an ideally completed) physics, but at the same time hold that the mentalistic

vocabulary we use to talk about intentional agents (i.e. persons) is indispensable and

irreducible to the physical descriptions available to us from the various core natural

sciences. Quine, inWord and Object , therefore calls the attribution of propositional

attitudes to ourselves and others an “essentially dramatic idiom”: mind is something we

thereby project into ourselves and others (WO 219). As unreal as the referents of such

idiomatic usages may be, however, Quine nonetheless regards this form of theater as

“humanly” and “practically” indispensable (WO 218, 221). But it is because the

attribution of intentionality is a fundamentally projective act that we are, in practice,

unsure where to draw the line: that is, there is no fact of the matter that would decisively

settle when such projections have gone too far. As Davidson put it, “we speak of the

‘intentions’ and ‘desires’ of simple animals” because we don’t know how else to talk

29 As Paul Churchland, one of the most prominent eliminative materialists, puts it: “FP [folk psychology] does indeed portray human cognition in terms of overtly sentential prototypes, viz. in terms of the many propositional attitudes. But there is no reason why it must becorrect in so representing our cognition, nor in representing itself in particular. Perhaps the internal kinematics and dynamics of human and animal cognition is not at all like the sentential dance portrayed in FP.... Perhaps we harbour instead a kinematics of activation patterns and a dynamics of vector-to-vector transformations driven by learned configurations of synaptic connections. Evidently it is not inconceivable that FP might someday be challenged by a better account of human nature. Evidently the process is already underway” (CPM 315). Dennett and other non-reductive physicalists would not fundamentally disagree with Churchland’s account of the physical processes that must underlie human cognition (to use a term that has become common in the philosophy of mind, Dennett and Davidson agree that the mental supervenemust on a physical base), yet they would disagree with Churchland’s notion that folk psychology could (and should) ever be superceded by a non-intentional, purely scientific, vocabulary. On supervenience, see: Jaegwon Kim, Mind in a Physical World: An Essay on the Mind-Body Problem and Mental Causation (Cambridge, MA: MIT Press, 1998).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. about them: “We have no better way to explain what they do” (SIO 128). Or as Quine

puts it:

Casting our real selves thus in unreal roles, we do not generally know how much reality to hold constant. Quandaries arise. But despite them we find ourselves attributing beliefs, wishes, and striving even to creatures lacking the power of speech, such is our dramatic virtuosity. We project ourselves even into what from his behavior we imagine a mouse’s state of mind to have been, and dramatize it as a belief, wish, or striving, verbalized as seems relevant and natural to us in the state thus feigned. In the strictest scientific spirit we can report all the behavior, verbal and otherwise, that may underlie our imputations of propositional attitudes, and we may go on to speculate as we please upon the causes and effects of this behavior; but, so long as we do not switch muses, the essentially dramatic idiom of propositional attitudes will find no place. (WO 219)

According to Quine’s view of mentality, in other words, the first objects of

anthropomorphism and the pathetic fallacy must needs be ourselves.

Adapting and extending Quine’s insights, Dennett argues that we may describe a

living human organism (any many other things besides) from three different

perspectives (or stances): the physical stance, the design stance, and the intentional

stance (Dennett’s version of Quine’s “dramatic idiom”). The physical stance simply

describes an object’s material make-up, while the design stance treats a thing

functionally: a physical description of a computer, for example, would account for all of

the materials that constitute its physical make-up, while a design description would

explain that pressing the “backspace” key will cause the cursor to move one space to the

left on the screen. The higher level intentional stance, however, treats entities like

humans (and sometimes even mice) as rational agents whose behavior can be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. characterized and predicted by means of intentional or men tali stic interpretations. But

along with Quine, Dennett does not believe there is any fact of the matter to which our

intentional idioms refer. The intentional states (such as ones of belief) postulated by the

intentional stance have no ontological existence: they are non-factual and only “appear”

when we assume the right attitude towards another human being. Dennett explains that

attributions of intentionality are

interpretations of the phenomena - a “heuristic overlay,” describing an inescapably idealized “real pattern.” Like such abstracta as centres of gravity and parallelograms of force, the beliefs and desires posited by the highest stance have no independent and concrete existence, and since this is the case, there would be no deeper facts that could settle the issue if - most improbably - rival intentional interpretations arose that did equally well at rationalizing the history of behavior of an entity. Quine’s thesis of the indeterminacy of radical translation carries all the way in, as the thesis of the indeterminacy of radical interpretation of mental states and processes. (CPM 239)

According to Baron-Cohen’s theory of autism, four different cognitive

subsystems or “mechanisms” make up what he terms the normal “mindreading system,”

which as a whole enables us to adopt the intentional stance towards others. The first is

called the “intentionality detector” (ID): “ID is a perceptual device that interprets

motion stimuli in terms of the primitive volitional mental states of goal and desire. I see

these as primitive mental states in that they are the basic ones that are needed in order to

be able to make sense of the universal movements of all animals: approach and

avoidance” (MB 32-3). In support of the existence of this subsystem Baron-Cohen cites

a number of studies: one (Reddy 1991) shows “that very young infants are sensitive to

changes in an adult’s goal. For example, they respond to the distinction between a give

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and a tease” (35).',0 Studies of lower primates have “identified cells in the temporal

lobe of the monkey brain that respond selectively to the sight of another animal facing

forward, even if seen in profile. One can think of these as part of ID, detecting the

animal’s goal (to move forward). Yet other cells fire selectively to tactile stimulation

by an agent other than oneself, which suggests that there may be specific neural

structures sensitive to another agent having a goal of doing something to the observer”

(MB 38);31 And in support of the subsystem’s localization and dissociability, he cites a

study that has shown that “some patients with focal brain damage have been found to

lose the specific ability to categorize things as animate or inanimate” (MB 38).32

The second mechanism in the mindreading system is what Baron-Cohen calls

the “eye-direction detector” (EDD). The EDD has three basic functions: “it detects the

presence of eyes or eye-like stimuli, it computes whether eyes are directed toward it or

toward something else, and it infers from its own case that if another organism’s eyes

are directed at something then that organism sees that thing” (MB 38-9). A number of

studies have shown that for the human perceptual system, eyes are the most important

part of the whole face. Daphne Maurer, for example, found “that 2-month-old infants

looked almost as long at the eyes as at a whole face but looked significantly less at other

parts of the face” (MB 39). Six months old have also been found to look two to three

30 Reddy, V., “Playing with other’s expectations: Teasing and mucking about in the first year”Natural in Theories of Mind, ed. A. Whiten. (Oxford: Blackwell, 1991). 31 Baron-Cohen cites: Perrett, D., and Mistlin, A. “Perception of facial characteristics by monkeys” in Comparative Perception, vol. 2: Complex Signals, ed. W. Stebbins and M. Berkeley (Wiley, 1990); Hietanen, J., and Perrett, D. “A role of expectation in visual and tactile processing within temporal cortex” inBrain Mechanisms of Perception and Memory: FromNeuron to Behavior, ed. T. Ono et al. (Oxford: OUP, 1991). 32 Warrington, E., and Shall ice, T. “Category specific semantic impairments” Brainin 107 (1984), pp. 829-854.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. times longer at a face looking at them than at a face looking away. In a study conducted

by Baron-Cohen and a colleague, they found that “computation of eye direction was

easily within the ability of normal 3-year-olds” (MB 40).33 Moreover, studies have

shown that we are particularly sensitive to mutual eye-contact: “there is clear evidence

of physiological arousal produced by mutual eye contact. For example, galvanic skin

responses increase with mutual eye contact, and brain-stem activation has been reported

in response to eye stimuli in monkeys” (MB 42).34 While physiological arousal can of

course be felt as both negative and positive in quality, in the case of human infants the

evidence suggests that the responses are usually pleasurable, “since eye contact reliably

triggers smiling” (MB 42).35 Indeed, it appears that the self-manipulation of arousal in

response to eye-stimuli is a very important part of an infant’s life:

Stem (1985)36 points out that an infant’s control over its visual system is precociously mature, enabling the infant to make or break eye contact and thus regulate the degree of eye contact and the amount of physiological arousal that the infant can cope with at a time. Too much might be uncomfortable; too little might be understimulating. Since what constitutes a comfortable level of arousal is likely to vary from one infant to another, it makes good sense that the infant should have its own regulatory mechanism to control this. There is mounting evidence that infants have a drive

33 Baron-Cohen, S., and Cross, P., “Reading the eyes: Evidence for the role of perception in the development of a theory of mind” inMind andLanguage 6 (1992), pp. 173-186. 34 Nichols, K., and Champness, B., “Eye gaze and the GSR” inJournal of Experimental Social Psychology (1971),1 pp. 623-626; Wada, 1., “Modification of conically induced responses in brainstem by shift of attention in monkeys” inScience 133 (1961), pp. 40-42. 33 Wolff, P., “Observations on the early development of smiling” Determinantsin of InfantBehavior , vol. 2, ed. B. Foss (Wiley, 1963); Stem, D., The First Relationship: Infantand Mother (Cambridge, MA: Harvard Univ. Press, 1977); Schaffer, H., “Early interactive development” Studiesin in Mother-Infant Interaction, ed. H. Schaffer (Academic Press, 1977). 36 Stem, D., The Interpersonal World of the Infant (Basic Books, 1985).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to maintain an optimal level of stimulation/' Furthermore, as all parents know, infants and toddlers love to play peekaboo, which is all about occluding the eyes and then revealing them. (MB 42)

According to Baron-Cohen, the intentionality and eye-direction detectors

together allow the infant to produce what he calls dyadic “representations” of various

states of affairs, such as: A wants food; B wants to open drawer; C sees me; D sees the

door. (Talk of “representations” is common in the cognitive scientific literature, but

such terminology should not be confused with representations in the philosophical

sense: there is no suggestion that these representations are in any way conscious or

linguistic in nature - rather, they are explicitly sub-personal and should be understood

as the “contents” of information-processing systems.) As Baron-Cohen explains, such

dyadic representations are obviously important but do not yet constitute the kind of

intersubjective cognitive relationships that would enable one “to be aware of a shared

universe,” the concept of an “objective reality” that Davidson argued can only be

produced by triangulation:

All these representations can be described as dyadic, since they only specify the intentional (i.e. mentalistic) relation between two objects (Agent and Object, or Agent and Self). Though that gets you pretty far, these mechanisms do not allow you to represent that you and someone else (whom we have been calling the Agent) are both attending to the same object or event. And yet that is exactly what one would need in order to be able to communicate about a shared reality and to feel that you and the other person are focusing on and thinking about the same thing. (MB 44)

37 Maurer, D., “Neonatal synaesthesia: Implications for the processing of speech and faces” in Developmental Neurocognition: Speech and Face Processingin the First Year of Life, ed. B. de Boysson- Bardies et a!. (Kluwer, 1993).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Joint attention, or triangulation, is the result of the third component in. Baron-

Cohen’ s theoretical model, which he calls the “shared-attention mechanism” (SAM):

“SAM’s key function is to build rather interesting things calledtriadic

representations.... Essentially, triadic representations specify the relations among an

Agent, the Self, and a (third) Object. (The Object can be another Agent, too.)” (MB 44-

5). As an example of a triadic representation, Baron-Cohen offers the following: “You

and I see that we are looking at the same object” (MB 45). Research findings suggest

that the shared-attention mechanism depends heavily (though not exclusively) upon our

perception of eye-direction and begins quite early in life: “gaze monitoring is seen in

infants from around 9 months of age, and which all children, the world over, show by

14 months or so. In this phenomenon, the infant turns in the same direction that another

person is looking at and then shows gaze alternation, checking back and forth a few

times to make sure (as it appears) that it and the other person are both looking at the

same thing, thus establishing shared visual attention on the same object” (MB 48).38

Moreover, children around the same time will begin to produce a “so-called

protodeclarative pointing gesture—that is, pointing with an outstretched index finger at

an object and then alternating the gaze again, checking back and forth a few times to

make sure (as it appears) that the other person has turned to look at the same thing the

toddler is looking at. This is a simple but effective way to direct someone else’s visual

38 Scaife, M., and Bruner, J., “The capacity for joint visual attention in the infant”Nature in 253 (1975), pp. 265-266; Butterworth, G., ‘The ontogeny and phyiogeny of joint visual attention”Natural in Theories o f Mind, ed. A. Whiten (Oxford: Blackwell, 1991).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. attention to a shared focal object” ( 4 8 ) : Such proto-declaratives constitute an. early,

pre-linguistic, form of ostensive communication that will flower into the wide range of

indexicals that are such an important part of mature language use. What makes the joint

attention that is the goal of proto-declarative behavior so important to the development

of an intentional stance is that it evinces interest not merely in the object itself but in the

mental attitude of another person towards that object as well. As Uta Frith puts it:

“Attending to the object that another person is also attending to is a major

developmental milestone, which presents the child with a multitude of learning

opportunities”(AEE 101). But the first thing such joint attention behavior shows is that

the infant has already grasped the existence of other minds as such: and the faces and

eyes of others appears to be the most natural places for infants to look for signs of

mentality.

Studies show that the eyes are particularly important. At a very young age,

children regard the eyes of others as providing privileged access to their mental states.

“When the goal of an action is uncertain,” experiments have demonstrated, “the first

place young children (and indeed adults) look for information to disambiguate the goal

is the person’s eyes” (MB 49).40 Moreover, it has also been shown that children use eye

direction as a way of detecting another agent’s goal. “When 3-4-year-olds were asked

Which chocolate will Charlie take?’ after being shown a display of four chocolates and

39 Bates, E., Benigni, L., Bretherton, I., Camaioni, K., and Volterra, V., “Cognition and communication from 9-13 months: Correlational findings” Thein Emergence o f Symbols: Cognition and Communication in Infancy, ed. E. Bates (Academic Press, 1979). 40 Phillips, W., Baron-Cohen, S., and Rutter, M., “The role of eye-contact in the detection of goals: Evidence from normal toddlers, and children with autism or mental handicap”Development in and Psychopathology 4 (1992), pp. 375-383.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Charlie’s face looking at one of these, they tended to pick the one he was looking at as

the goal of his next action” (MB 49). In a variation on this study, the face of Charlie

was again depicted as looking at one of the four sweets and the 3 to 4 year olds were

asked “Which one does Charlie want?” As Baron-Cohen reports, “children of this age

had no difficulty at all in inferring Charlie’s desire... from his eye direction. This was

particularly striking, as in a retest of this experiment the display also included a

distracter cue: a big, bold, black arrow pointing at another of the four chocolates.

Normal 3-4-year-olds appeared to ignore this ‘unnatural’ cue, and predominantly used

the ‘natural’ cue of eye direction” to determine Charlie’s mental state (MB 49-50).

The research of Baron-Cohen and his colleagues has shown that the

“intentionality detector” and the “eye-direction detector” are intact and normally

functioning in children with autism. However, it has been shown that autistic children

suffer a massive impairment in the functioning of the shared-attention mechanism, and

therefore do not seem capable of social triangulation: “Children with autism often do

not show any of the main forms of joint-attention behavior. Thus, they do not show

gaze monitoring,41 nor do they show the related behaviors of attempting to direct the

visual attention of others by using the pointing gesture in its ‘protodeclarative’ form”42

41 Mundy, P., Sigman, M., Ungerer, J., and Sherman, T., “Defining the social deficits in autism: The contribution of nonverbal communication measures” inJournal of Child Psychology andPsychiatry 27 (1986), pp. 657-669; Loveland, K., and Landry, S., “Joint attention and language in autism and developmental language delay” inJournal of Autism andDevelopmental Disorders 16 (1986), pp. 335- 349. 42 Baron-Cohen, S., “Perceptual role-taking and protodeclarative pointing in autism”British in Journal of DevelopmentalPsychology 1 (1989), pp. 113-127; Mundy et al. “Defining the social deficits in autism: The contribution of nonverbal communication measures” (1986); Curcio, F., “Sensorimotor functioning and communication in mute autistic children” Journalin o f Autismand Childhood Schizophrenia 8 (1978), pp. 281-292.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (MB 66). Interestingly, this is not because they cannot point at all. Baron-Cohen

reports that “they do use the pointing gesture for some other, non-joint attentional

functions, such as to request objects that are out of reach” (66).4j The crucial difference

between “proto-imperative” and “proto-declarative” pointing behavior is that the former

treats other agents in a purely behavioral goal-directed way (i.e. “give me that!”) while

the latter is directed at affecting the mental states of other persons: “look at that!”

When these three subsystems are working properly, they lead to the

development of a fourth component that finally provides a full-blown intentional stance:

Baron-Cohen calls this final component the “theory-of-mind mechanism” (ToMM).44

(The use of the word “theory” in this phrase can be misleading, for it is not meant to

suggest anything consciously grasped: it is rather like Davidson’s “commonsense

apparatus” or Dennett’s “folk psychology”.) The theory-of-mind mechanism is what

allows us to so easily and intuitively “mind-read” other humans, regardless of the fact

that “mind” itself is clearly not a fact that is visible in the way that our bodies or

behavior are. Along with the “shared attention mechanism,” this cognitive subsystem is

either missing or severely impaired in children who suffer from autism: thus, as it is

commonly put in the scientific literature, they lack an intuitive “theory of mind,”

rendering them “mind-blind.” According to Baron-Cohen, the “theory-of-mind

mechanism” utilizes the output of the other three in order to enable one to grasp the full

range of “epistemic mental states (which include pretending, thinking, knowing,

4j Baron-Cohen, S., “Perceptual role-taking and protodeclarative pointing in autism” (1989). 44 Baron-Cohen notes that he is taking the phrase “theory-of-mind” from Leslie, A., “ToMM, ToBy, and Agency: Core architecture and domain specificity” inMapping the Mind: Domain Specificity in Cognition and Culture, eds. L. Hirschfeld and S. Gelman (Cambridge: Cambridge Univ. Press, 1994).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. believing, imagining, dreaming, guessing, and deceiving)” and to tie them, together

“into a coherent understanding of how mental states and actions are related” (MB 51).

Interestingly, Leslie has argued that the earliest and most important epistemic mental

state for the development of a full-blown theory of mind pretence,is which underlies

the spontaneous pretend play that is common among very young children but which has

been shown to be absent in those with autism. Leslie finds the early onset of the ability

to pretend remarkable:

Pretending ought to strike the cognitive psychologist as a very odd sort of ability. After all, from an evolutionary point of view, there ought to be a high premium on the veridicality of cognitive processes. The perceiving, thinking organism ought, as far as possible, to get things right. Yet pretense flies in the face of this fundamental principle. In pretence we deliberately distort reality. How odd then that this ability is not the sober culmination of intellectual development but instead makes its appearance playfully and precociously at the very beginning of childhood.... How is it possible for a child to think about a banana as if it were a telephone, a lump of plastic as if it were alive, or an empty soap dish as if it contained soap?45

According to Leslie, the reason pretence is so important, along with the reason it is so

essential that it appear this early in childhood development (usually around 18 months),

is that pretend play involves the same distinction between prepositional attitudes and

prepositional contents that enables subsequent attributions of mental states (such as

belief) to other persons. That is, before the full-blown acquisition of language and the

web of propositional attitudes as such, pretend playenacts the logical distinction

43 Alan Leslie, “Pretence and representation: The origins of ‘theory of mind’”Psychological in Review 94 (1987), pp. 412-426, p. 412. Cited in AEE 81.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between prepositional attitudes and contents that constitute the building blocks of the

space of reasons.

In pretend play, a three-way relationship is established, between an agent

(usually the child), a primary representation (the actual object being played with) and a

secondary representation that is de-eoupied from the primary one (which represents the

content of the pretence). Thus a child can pretend that a banana is a telephone while

remaining aware that it is a banana as well: as Karmiloff-Smith and Russell explain,

“the primary and decoupled representations involve different and separate levels of

processing and obey distinct causal and logical inferential constraints” (CPM 254).

Reality, in other words, is no longer identical with what someone may think about it.

As Frith explains, once this decoupling is accomplished, the child can go on to attribute

thoughts to others that are distinct from reality as well: “When decoupled, the

representation are no longer copies of the real world. Therefore, they can be attached to

a person’s wishes, thoughts, or memories. For instance, the child understands: mother

thinks about the banana It’s‘ a telephone,'' and laughs” (AEE 81). This is an absolutely

crucial step in the development of the child’s “theory of mind.” Once the innate ability

to decouple is activated, the attribution of prepositional attitudes such as belief towards

others becomes common. As Frith notes: “Given sufficient learning experience, the

mentalizing mechanism enables the child to leam surprisingly fast about beliefs and

deception” (AEE 81).

Leslie’s theory of pretence therefore predicts that children with autism, who do

not exhibit spontaneous pretend play, should not be able to grasp the concept of belief,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. since the understanding of this prepositional attitude is developmental!}' dependent

upon the earlier mastery of the ability to pretend. Dennett has argued that the best way

to test if a child has a concept of belief is to see if the child understands that another

person might hold a false belief: “this,” explains Baron-Cohen, “might constitute a

litmus test of whether an organism had a ‘theory of mind,’ in that in such cases it

becomes possible to distinguish unambiguously between the child’s (true) belief and the

child’s awareness of someone else’s different (false) belief’ (MB 69-70). Following

Dennett’s suggestion, Baron-Cohen, Leslie and Frith devised an experiment using two

dolls (named Sally and Anne) in order to test the understanding of belief among three

different child populations: children with autism, children with Down’s syndrome and

normal children. Frith describes the “Sally-Anne Experiment” this way:

Sally has a basket and Anne has a box. Sally has a marble and she puts it into her basket. She then goes out. Anne takes out Sally’s marble and puts it into her box while Sally is away. Now Sally comes back and wants to play with her marble. At this point we ask the critical question: “Where will Sally look for her marble?” The answer is, of course, “In the basket.” This answer is correct because Sally has put the marble into the basket and has not seen it being moved. Shebelieves the marble is still where she put it. Therefore she will look in the basket even though the marble is not there any more. (AEE 83-84)

As Frith remarks, the “results were striking”: most of the normal children and children

with Down syndrome correctly answered the question “Where will Sally look for her

marble?” In contrast, nearly all of the children with autism pointed to the box. (The

results were later replicated using live humans in the roles of Sally and Anne.) In

accordance with the predictions of Leslie’s theory of the developmental role of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. pretence, they were unable to infer from the situation that Sally would have a false

belief about the location of her marble: in other words, they showed that they had no

intuitive grasp of the concept of belief and thus could not see that another person’s

thoughts might be different from their own. Without prior mastery of the ability to

decouple representations from reality through pretence, the autistic children were

unable to attribute the prepositional attitude of belief to Sally: they were blind to the

otherness of Sally’s mind.

In a negative way, this confirms Davidson’s point that surprise is criterial of

rationality. As we already noted, Davidson holds that “surprise about some things is a

necessary and sufficient condition of thought in general,” for it demonstrates that the -

surprised person grasps the distinction between “what is believed and what is the case”

(SIO 96). Here, however, it seem that we need to add to Davidson’s criterion (though

perhaps it is already implicit in it) the requirement that one be able to understand the

possibility of another person experiencing surprise. Interestingly, Frith notes that the

anticipation and observation of surprise in others are important and regular aspects of

children’s entertainments (as, of course, are deception and disguise, which also

presuppose the ability to understand that other people might hold false beliefs):

A common feature in children’s stories and puppet plays all over the world is the surprise created by one actor knowing something that another actor does not. For instance, Punch is hit over the head by Judy. This is funny because, unbeknown to Punch, Judy was coming up behind him with a rolling pin. Children from about age three watch such shows with every sign of apparent pleasure and anticipation. They know that something that a person has not seen cannot be in the mind of that person, and will therefore cause a surprise. If a child can implicitly take account

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of another person’s thoughts, this kind of inference will be obvious. (AEE 82)

Sadly, it is this very “obvious” kind of inference that children with autism are blocked

from readily seeing, as the Sally-Anne experiment clearly demonstrated.

The studies and analyses of autistic mindblindness by Baron-Cohen and his

colleagues foreground what kinds of cognitive capacities are taken for granted in the

theories of philosophers such as Quine and Davidson, while at the same time

confirming that empathy (Quine) and triangulation (Davidson) are absolutely necessary

aspects of normal human cognitive development. Indeed, it appears that the

fundamental role of pretence in our ability to grasp the mental states of another person

obliquely supports Quine’s treatment of the attribution of propositional attitudes as a

kind of dramatic idiom: without the ability to pretend, we are blind to the existence of

other minds (otherbehavioral entities will be perceptible, for sure, but those entities

will not be endowed with the full range of propositional attitudes absent the prior

mastery of pretence). Quine, it turns out, is also absolutely right when he suggests that

empathy plays a crucial role in language acquisition: Baron-Cohen’s findings about the

centrality of the eyes to the establishment of joint attention corroborates Quine’s

intuitive analogy between empathetic response and facial recognition. Indeed, it seems

that mindblindness is intimately related to abnormalities in facial recognition and

perception. As Frith reports, “experiments have shown that people with autism cannot

remember faces as well as they can remember buildings or landscapes”46 and a

46 Blair, R.J.R., Frith, U., Smith, N., Abell, F., andCipolotti, L., “Fractionation of visual memory: Agency detection and its impairment in autism”Neuropsychologia in 40 (200), pp. 108-118.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. “neuropilysiological study by Robert Shultz and colleagues at Yale has demonstrated

that the part of the brain that is normally specialized for faces is not specialized in

autistic people”47 (AEE 104). In fact, the studies of Shultz showed that when

processing faces the brains of autistic subjects showed activity in the same areas

activated by the visual processing of nonface objects in normal control subjects (AEE

204). For people with autism, it appears that there does not seem to be anything

particularly special about the human face.

The inattentiveness of autistics to the appearance of mentality in the face

negatively affects their ability to team language, consistent with Quine’s correlation of

empathy with language acquisition. Studies of language learning by the developmental

scientist Paul Bloom show that a meeting of minds by means of a meeting of the eyes

JO plays a crucial role in normal language learning. As Frith describes Bloom’s findings:

One old idea is that the sound of a word has to be paired with the sight of a stimulus, and that from such pairings lasting representations are formed in memory. According to Bloom such a mechanism will not be sufficient. It would lead to frequent incorrect pairings of word sound and word meaning, which are actually quite rare in young children’s language. Consider the novel word “fendle” spoken when looking at a novel object, an unusual red ball with yellow spikes. Does “fendle” refer to the whole object, to the yellow spikes, or perhaps to one particular spike? If you follow the speaker’s gaze, and if you take into account what they said before (e.g., “Here is a strange one”), you would get clues to what the speaker meant (e.g., “fendle” refers to a red ball with yellow spikes). To understand the speaker’s intention, the intuitive mentalizing mechanism... will play a role. The 18-months revolution, which may coincide with a leap in the

47 Schultz, R.T., Gauthier, I., Klin, A., et al., “Abnormal ventral temporal cortical activity during face discrimination among individuals with autism and Asperger’s syndrome”Archives in of General Psychiatry 57 (2000), pp. 331-340. 48 Paul Bloom, How Children Lean the Meaning o f Words (Cambridge, MA: MIT Press, 2000).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. maturation of the brain system that underpins mentalizing, also coincides with a leap in word learning. Bloom argues that this is more than coincidence. (AEE 120)

Consistent with the theories of mind-reading and mind-blindness proposed by Baron-

Cohen and others, Bloom has shown that the direction of the adult’s gaze plays a crucial

role in the child’s acquisition of words: after all, we have already noted that the first

place people generally look in order to detect another person’s goal or mental state is

his or her eyes. It is unsurprising, therefore, that autistic children, while showing

themselves capable of learning language, acquire it in ways radically different from

normal children. In one study, for example, both normal and autistic children were

individually placed in a room in which no other person was present. Instead, a voice

was pumped in from a speaker above, saying “fendle” at the same time as the child

touched a novel object. As Frith explains, “this situation has all the prerequisites for

learning a sound and sight association, but not the prerequisites for intention tracking,

as there is only a disembodied voice” (AEE 120). Consistent with the mindblindness

explanation, the children with autism were able to learn the novel word under these

conditions while the normally developing children did not. As Frith comments, “While

normal children ignored the sound from above when no speaker was present, so autistic

children may well ignore the subtle cues to intentions when face to face with a speaker”

(AEE 120-121). Of course, as we noted earlier, it makes no sense to suggest, as Quine

does, that in a normal language-learning situation, empathy is exercised by one or

another of the parties: clearly both teacher and student, adult and child, native and

linguist need to be empathetic for effective language training to occur. Bloom agrees:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. he suggests that “given the ability to attribute intentions and other mental states to

others, it is after all not such a mystery that young normal children manage to acquire

language without much trial and error. It is not only the children, but the other people

around them, who continuously monitor mental states.... When a child leams the word

for ‘pudding,’ it is most likely that the word will have been uttered at exactly the right

moment so that the child does not mistakenly learn to use the word ‘pudding’ to mean

‘raisin,’ or ‘yellow,’ or ‘stop it’” (AEE 121). So when we read the literary critic Paul de

Man declare that “Man can address and face other men... because he has a face,” we can

nod and agree.49 But when he goes on to add that “he has a face only because he

partakes of a mode of discourse that is neither entirely natural nor entirely human,” we

can understand that he is almost exactly wrong. Whatever “modes of discourse” we

may be able to partake of, we can do so only because we always already (as the saying

goes) perceive the significance of the human face beforehand: the meaningfulness of the

face lies at the cognitive ground of our linguistic capacities and is certainly not the mere

“product” of any figures of speech.50 By arguing that self and interpersonal

communication are reducible to the autonomous workings of “language” (whatever that

49 Paul de Man, The Rhetoric o f Romanticism (New York: Columbia Univ. Press, 1984), p. 90. 50 For example, prosopopoeia. On de Man’s view of the rhetorical trope of prosopopoeia, see esp. “Autobiography As De-Facement” and “Shelley Disfigured” inThe Rhetoric of Romanticism, pp. 67-81 and pp. 93-123. In “Shelley Disfigured,” for example, we are told that the human face (along with the sense of subjectivity the figure of the “face” is said rhetorically to produce) endlessly emerges and dissolves within a tropological play of face-giving and defacement, which he calls “prosopopoeia” (from “prosopon poien, to confer a mask or a face”). For de Man, there is no end to prosopopoeia for it is the very “madness of words” and “no degree of knowledge” can stop it (122). Incurable, the only therapy possible for this condition is the ironic insight that there is no point celebrating or denouncing it, since we readers and subjects “are its product rather than its agent” (122).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. word might mean abstracted from the intentions of humans who actuallyuse words), de

Man has put matters entirely backwards.

• The Nature of Reason: Brains and Minds, Syntax and Semantics

As I hope is evident, the study of autism has added immensely to our

understanding of the crucial role played by the human face and facial perception in the

normal development of the human mind, including our ability to acquire, to recall

Davidson’s phrase, “the gift of tongues.” If the intentional stance is, as Dennett likes to

say, a kind of pattern recognition, it seems that one of the most important perceptual

patterns may be that of the human face (cf. Dennett’s “Real Patterns”51). To return full

circle to the Aristotelian view of humans as “rational animals” with which we began,

developmental psychology so far bears out our contention that faces do play a special

role in the cognitive transformations that bridge the animal and the rational sides of our

being, while at the same time making clear that, of course, noreal chasm exists between

these two terms. In a way, the conceit of a bridge is an idle one: as developmental

psychology and cognitive neuroscience (often working in concert) continue to show,

51 Daniel Dennett, “Real Patterns” inBrainstorms: Essays on Designing Minds (Cambridge, MA: MIT Press, 1998), pp. 95-120. On p. 98 we find: “Where utter pattemlessness or randomness prevails, nothing is predictable. The success of folk-psychological prediction, like the success of any prediction, depends on there being some order or pattern in the world to exploit. Exactly where in the world does this pattern exist? What is the patternof! Some have thought, with Fodor, that the pattern of belief must in the end be a pattern of structures in the brain, formulae written in the language of thought. Where else could it be? Gibsonians might say the pattern is ‘in the light’—and Quinians (such as Davidson and I) could almost agree: the pattern is discernible in agents’ (observable) behavior when we subject it to ‘radical interpretation’ (Davidson) ‘from the intentional stance’ (Dennett).”

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rationality is inextricable from the particular biological nature of the human organism.

The development of reason, while wondrous, is nonetheless utterly natural.

However, while it seems .reasonable to assume that no one has come across a

rational being that is not at the same time an animal as well, there are aspects of

rationality that seem to resist reduction tomere biology. Descartes’ untenable

distinction between the mental(res cogitans) and the corporeal(res extensd) is only the

most exaggerated form that the perceived antinomy between reason and nature has

taken in the philosophical tradition. But even to many anti-Cartesian philosophers, the

web of propositional attitudes that constitute the space of reasons seems clearlysui

generis when judged against the kind of law-like causal regularities that structure the

space of nature. This was Davidson’s point when he remarked that “whichever way we

talk, there remains the difference, with respect to rationality, between the infant and the

snail on the one hand, and the normal adult person on the other” (SIO 95). For

Davidson, the normative—that is, rational—inter-relationships that characterize the

propositional attitudes “have no echo in physical theory” and therefore cannot be

reduced to the kinds of empirical descriptions available from the natural sciences.52

This is consistent with Wilfrid Sellars’ point that since mental states are inherently

normative, descriptions of them cannot be characterized as empirical: “In characterizing

an episode or a state as that ofknowing, we are not giving an empirical description of

that episode or state; we are placing it in the logical space of reasons, of justifying and

32 Donald Davidson,Essays on Actions and Events (Oxford: Clarendon Press, 1980), p. 231.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. being able to justify what one says,”5'1 Sellars' point is that mental states or episodes

are not law-bound as are phenomena in the space of nature. John McDowell, who has

made Sellars’ distinction between the space of reasons and the space of nature the

conceptual crux of his own philosophical views, explains the distinction this way:

What would the logical space of nature be? I think we capture the essentials of Sellars’s thinking if we take it that the logical space of nature is the logical space in which the natural sciences function, as we have been enabled to conceive them by a well- charted, and in itself admirable, development of modem thought. We might say that to place something in nature on the relevant conception, as contrasted with placing it in the logical space of reasons, is to situate it in the realm of law. But what matters for Sellars’s point is not that or any other positive characterization, but the negative claim: whatever the relations are that constitute the logical space of nature, they are different in kind from the normative relations that constitute the logical space of reasons. The relations that constitute the logical space of nature, on the relevant conception, do not include relations such as one thing’s being warranted, or—for the general case—correct, in the light of another. This is what Sellars is saying when he insists that “empirical description” cannot amount to placing something in the logical space of reasons. (MW xv)54

Sellars’ point, surely correct, poses a serious challenge to the integration of scientific

research (especially that of contemporary cognitive neuroscience) with philosophical

reflection. After all, the distinction Sellars and McDowell are insisting on can easily be

interpreted as one between the brain as a physical object studied by the natural sciences

33 Wilfrid Sellars, “Empiricism and the Philosophy of Mind,” in Herbert Feigl and Michael Scriven, eds., Minnesota Studies in the Philosophy o f Science, vol. 1 (Minneapolis: University of Minnesota Press, 1956), pp. 253-329, pp. 298-9. 54 John McDowell, Mind and World (Cambridge, MA: Harvard Univ. Press, 1994). Reference to this text will be given parenthetically with the abbreviation MW.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and the mind as a rational consciousness understandable only according to what

Davidson likes to call the “constitutive ideal of rationality.”55

Indeed, McDowell has tackled this very issue in a critique of Dennett’s essay

“Toward a Cognitive Theory of Consciousness”: as its title suggests, Dennett’s piece

attempts to explain consciousness by means of the kinds of sub-personal cognitive

mechanisms that play such an important explanatory role in the theories of Baron-

Cohen and others working at the forefront of contemporary science of mind.56

McDowell has no problem with cognitive science as such: he disavows any “know-

nothing refusal to acknowledge the rich promise of cognitive science” (MVR 356).

However, he believes that we must be careful when interpreting and articulating the

relationship between a person’s mind and the cognitive mechanisms that play an

enabling causal role in the biological existence of that mind: that is, between two levels

of being and explanation that McDowell calls the “personal” and the “sub-personal.”

The sub-personal cognitive systems, such as the clustered populations of face cells that

Martha Farah hypothesized make up a distributed neural network geared toward the

recognition of particular faces, belong, according to McDowell, to the space of nature:

they are bound, as are all physical matter and energy, by causal laws. Such face cells

may, and clearly do, ultimately enable a rich phenomenological experience of the faces

of family, friends and strangers, yet the full-blown conscious experience of facial

55 This phrase comes from Davidson’s “Mental Events” inEssays on Actions and Events, p. 223. 36 Daniel Dennett, “Towards a Cognitive Theory of Consciousness”Brainstorms: in Philosophical Essays on Mind and Psychology (Montgomery, VT: Bradford Books, 1978), pp. 149-173. McDowell’s critique is called “The Content of Perceptual Experience” in John McDowell,Mind, Value and Reality (Cambridge, MA: Harvard Univ. Press, 1998), pp. 341-358. References toMind, Value and Reality will be given parenthetically with the abbreviation MVR.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. recognition and perception is distinct in kind, and not merely in degree, from the

information-processing that goes on at the sub-personal levels in various regions of the

brain. McDowell’s fundamental point is that a wholeperson recognize the faces of

others, not parts of that person’sbrain: to identify the brain with the person is to

conceptually confuse radically distinct levels of description. The brain is a physical part

of the space of nature, while the person lives and thinks in the space of reasons.

However, this point is not fundamentally incompatible with Dennett’s own (though

Dennett does open himself up to criticism by occasionally blurring the distinction, but I

think this is often simply due to rhetorical and conceptual enthusiasm). Intentional

states, according to Dennett are after all only attributable to whole “systems,” not parts

or subsystems. As Andrew Brook and Don Ross explain: “...intentional states are not

ascribed to parts of a person, even parts as large as the brain. Intentional states are

states of the whole person, indeed a person/environment whole. If intentional states are

complex triangulations of behavior/brain/environment interactions, then they cannot be

reduced to brain-states.”57

Another way of getting atthe distinction that McDowell is keen to bring forth is

by thinking of the gap between syntax and semantics. In fact, Dennett himself is the

one who first articulated the conceptual distinction between brains and persons this

way: brains and their parts, according to Dennett, are syntactic engines, while whole

persons are semantic ones. Sub-personal information-processing systems that make up

37 Andrew Brook and Don Ross, eds.,Daniel Dennett (Cambridge: Cambridge Univ. Press, 2002), p. 18.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the structure of the brain are physical in nature, and thus they do not know the first thing

about “meanings.” As McDowell puts it:

A sub-personal... informational system is a physical mechanism, connected to its surroundings by transducers that convert physical impacts from outside into events of the sort that the system can work on, and perhaps by transducers that convert the system’s end products into physical interventions in the exterior. The system knows nothing even about the character of the immediate physical impacts on the input transducers, or the immediate physical interventions in the exterior that result from its operations by way of the output transducers, let alone about the nature and layout of the distal environment. The operations of the system are determined by structures exemplified in the initial contributions of the transducers, and in intermediate events and states in the system, which have nomeaning for the system. In short, in Dennett’s own memorable and exactly right phrase, the system is a syntactic engine, not a semantic engine. (MYR 350- 351, emphasis added)

Because the operations of the brain are fully determined by its material structure along

with the input it receives from the environment, it cannot be said to “know” or

“understand” or “communicate” anything, even to the mind of the person who possesses

that very brain. That is to say, McDowell goes on, “we have inside us something that is

not intelligent at all (it knows nothing and understands nothing)” (MVR 353). And yet

clearly, “That is not the truth about us” (MVR 354). We, as persons, understand, know,

communicate, reason, perceive, recognize and so on. Our lives are meaningful through

and through. Structure never, in and of itself, produces meaning (and this, of course, is

what is fundamentally wrong about all structuralist theories of language: they account

for the abstract inter-relationships of sign-systems, but are incapable of explaining how7

signs canmean something to language users). When we recognize the familiar face of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. another person standing before us, for example, we certainly can do so only because an

array of sub-personal information-processing systems participate in and enable the

accomplishment of that everyday cognitive task, but for the whole person, facial

recognition is simply the recognition of the presence of a particular person whom we

know. The syntactic workings of the cognitive subsystems enable, but do not

constitute, the meaningful experience of interpersonal recognition. As McDowell

explains:

At the level of internal machinery, it is useful to talk of sensory systems as information-processing devices; but for the animal its sensory systems are modes of openness to features of its environment. Information-processing characterizations of the internal machinery figure in explanations of how it can be that animals are in touch with their environments. The “as if ’ content that is usefully deployed at the lower level helps make intelligible the genuine content that appears at the higher level by way of “enabling” explanations, not as somehow constituting that content. Since there is no getting around the fact that the internal machinery is really only a syntactic engine, the attempt to see a constitutive relation between the lower and the upper levels undermines our hold on the fact that animals are semantic engines. (MVR 354-354)

We might put the consequences of McDowell’s point this way: if Dennett (following

Quine) had already acknowledged that there was no fact of the matter which could

ground our attribution of a particular propositional attitude toanother person, then

perhaps there is no fact of the matter (in our own brain) which constitutes our ability to

attribute propositional attitudes at all.

Consciousness, in other words, is still unexplained. As McDowell concludes,

“consciousness itself escapes Dennett’s cognitivistic net; he offers what may be an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. enabling explanation of consciousness, but not a constitutive one. And in one sense,

this leaves us without an account of consciousness. We lack an account of what it is,

even if we have an account of what enables it” (MVR 357). For McDowell, this does

not mean, however, that consciousness is now relegated to an immateriaiist ontology:

“as if, in denying that consciousness is a matter of configurations in the satisfyingly

material medium of the nervous system, we were committed to regarding it as a matter

of configurations in an immaterial medium instead” (MVR 357). It seems to me that

McDowell’s conclusion here, while on the one hand providing a salutary check to

certain tendencies in Dennett’s thought that would problematicallyidentify sub-personal

cognitive structures with consciousness, on the other hand supports Dennett’s more

fundamental point that mentality is not factual in nature, and is simply the result of the

intentional stance we take towards human entities, both ourselves and others. To

confirm McDowell’s point in a spirit with which I suspect he would not agree: mind, we

can agree, is not a fact of nature, even if there is nothing more natural in the world.

The outcome of these reflections, however, is that we are left without a

satisfying account of how we perceive meaning and mentality in the faces of other

humans. We have learned a great deal from the scientific literature about the structural

reasons why the face would be a perceptual object of singular importance to the human

organism: but the way we grasp mentality in the face of another is still unclear. After

all, as Karmiloff-Smith and Russell note, “there is nothing specificallymental about

human faces” (CPM 253). This is certainly so because there is nothing specifically

mental about anything at all. It is evident from the development of this argument that

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. cognitive science alone—while full of tantalizing clues and immensely enlightening

about the underlying cognitive structures—cannot provide a full account of the rich

semantics of the human face. For this, we will have to turn elsewhere. Luckily, as we

have already noted, Wittgenstein found facial perception (and the related art of

physiognomy) profoundly significant and his later philosophical writings tell us a great

deal about how we perceive meaning and mentality in the faces of other people.

Considering the great and wide-ranging importance he attributed to the ancient art of

face reading, we could even call his whole later philosophy a form of conceptual

physiognomy: as he himself rather cryptically noted at one point in theInvestigations,

“Meaning is a physiognomy” (PI §568). Having now reached the present limit of

scientific explanation (though who knows what the future will bring?), we can turn in

the following chapter to Wittgenstein’s philosophical reflections on the meaning of the

face.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2. Wittgenstein’s Faceof Meaning

Wittgenstein’s interest in faces and physiognomy can be traced to his early and

enthusiastic reading of Otto Weininger’sSex and Character, a treatise on the nature of

sexual difference that we know Wittgenstein highly esteemed.58 Weininger regarded

physiognomy as possessing great methodological importance, since he understood it to

study the visible manifestation of the parallelism between the psychic character and the

physical morphology of a given man or woman: “the parallelism between

characterology and morphology in the widest application,” Weininger wrote, “makes us

look forward to the time when physiognomy will take its honourable place amongst the

sciences...” (SaC 36). Wittgenstein was intimately familiar with this text, and so it is

unsurprising that elements of its discussion of faces would find their way into his later

physiognomic remarks. For example, the termaspect, which appears in Weininger’s

comments on the facial expressions of men of genius (“the number of different aspects

that the face of a man has assumed may be taken almost as a physiognomical measure

of his talent” [SaC 65]), would later come to constitute the subject of the longest section

of Part II of the Philosophical Investigations (sec. xi: on seeing aspects, which also

happens to include the discussion of the “physiognomy of a word”). That this

terminological echo is not unintended is suggested by the fact that in earlier versions of

58 Otto Weininger, Sex and Character, trans. of 6th German ed. (London: William Heinemann, 1906). References to this text will be given parenthetically with the abbreviation SaC. On Weininger’s influence on Wittgenstein, see Ray Monk’s excellent biography,Ludwig Wittgenstein: the Duty o f Genius (New York: Penguin, 1990). References to Monk’s biography will be given parenthetically with the abbreviation WDG. I am indebted to Monk’s text for drawing to my attention the connections between Weininger’s, Spengler’s and Wittgenstein’s ideas.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the remarks that would eventually become §§536-537 of the Investigations ,

Wittgenstein explicitly cites Weininger in conjunction with the idea of an “aspect of the

face itself’ (PI §536).39 Moreover, this physiognomic association helps to explain why

Wittgenstein’s initial definition of the concept of “noticing an aspect” would make

reference to an experience of facial perception: “I contemplate a face, and then suddenly

notice its likeness to another. Isee that it has not changed; and yet I see it differently. I

call this experience ‘noticing an aspect’” (PI p. 165). But while Wittgenstein’s interest

in aspects appears to be indebted to his reading of Weininger, the conceptual use he

makes of the term (which we will turn to later) is both stranger and broader than the

relatively ordinary use it has inSex and Character. On the whole, in fact, it appears

that aside from an occasional and loose terminological borrowing, the most significant

lesson Wittgenstein learned from Weininger’s physiognomy was simply the importance

of physiognomy itself, especially its fundamental belief in the expressive

meaningfulness of the human face. For example, commenting upon a timid face (in the

passage that in earlier versions cites Weininger), he writes: “the timidity does not seem

to be merely associated, outwardly connected, with the face; but fear is there, alive, in

the features. If the features change slightly, we can speak of a corresponding change in

the fear” (PI §537).

In addition to absorbing a physiognomic regard for the face from Weininger’s

Sex and Character, the later Wittgenstein, inspired by Oswald Spengler’sThe Decline

59 For example, we find in the manuscript text now available as Philosophicalthe Remarks: “Wenn ich sage, dieses Gesicht hat den Ausdruck der Milde, Giite, Feigheit, so scheme ich nicht nur zu meinen daB wir die und die Gefiihle mit dem Anblick des Gesichts assoziieren, sondem ich bin versucht zu sagen, daB das Gesicht ein Aspekt der Feigheit, Giite, etc., selbst ist. (Vergleiche z.B. Weininger.)”

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the West, would come to perceive in the art of physiognomy conceptual implications

reaching far beyond its original focus on the face.60The Decline of the West is a work

of cultural historiography that conceives of cultures as cyclical and organic in nature,

each having its own individual character and destiny: “Each Culture has its own

possibilities of self-expression which arise, ripen, decay, and never return. There is not

one sculpture, one painting, one mathematics, but many. Each is in its deepest essence

different from the others, each limited in duration and self-contained” (DoW 17).

According to Spengler, cultures are not susceptible of scientific interpretation, since the

methods of science grasp nothing but the laws of causality. Destiny, not causality, is

what structures the history of cultural forms (as it does the teleological life-cycles of

organisms) and so the proper method of cultural study will not be scientific, but

morphological, or as Spengler preferred to put it,physiognomic: “The Morphology of

the organic, of history and life and all that bears the sign of direction and destiny, is

called Physiognomic” (DoW 71). In keeping with this characterization of the

historian’s methodology, Spengler called the capacity to perceive the morphological

destiny of a culture a “physiognomic flair,” about which Helmut Wemer, the editor of

the 1959 abridgement of the Decline, offered the following explanation: “The word

‘destiny’ expresses an indescribable inward certainty: causality carries the notion of

law. The physiognomic flair, by which it is possible to read a lifetime, a fate, from a

face, operates without deliberate effort or any system. It is far removed from cause and

effect” (DoW 76). Discussing the study of economics, Spengler wrote: “Economics has

60 Oswald Spengler,The Decline o f the West (1918). References to this text will be given parenthetically with the abbreviation DoW.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. no system, but a physiognomy. To fathom the secret of its inner form, itssoul, demands

the physiognomic flair” (DoW 399). Since cultures are not structured by the

mechanistic laws of causality, attempts at causal explanation are misguided. An

entirely novel methodology is required to see and describe the morphological destiny

that each culture obeys, and Spengler can think of no better way to describe this form of

seeing than the analogy of physiognomy provides: we see the form of a culture’s

destiny in the same way we see the truth revealed by a person’s face. Physiognomy is

the key to morphological analysis.

Wittgenstein found this methodological orientation compelling, and it strongly

influenced the development of his later descriptive, anti-theoretical philosophical views.

For if the later Wittgenstein found theoretical explanations of philosophical problems to

be profitless (“For me”, he said in 1930, “a theory is without value. A theory gives me

nothing”), morphological description seemed a viable methodological alternative:

“What I give,” he said in one lecture, “is the morphology of the use of an expression”

(WDG 304, 303). This later orientation is forcefully expressed in his critical response

to Frazer’s The Golden Bough. Wittgenstein reacted with great hostility to Frazer’s

rationalistic explanations of “primitive” rituals as mistaken, proto-scientific

experiments. Rather than attempt toexplain the meaning of the rituals, which

Wittgenstein regarded as foolish, he suggested (invoking Spengler) that they should

simply be described, orshown, in a “perspicuous presentation”

...by arranging the factual material so that we can easily pass from one part to another and have a clear view of it - showing it in a perspicuous way. For us the conception of a perspicuous

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. presentation is fundamental. It indicates the form in which we write of things, the way in which we see things. (A kind of Weltanschauung that seems to be typical of our time. Spengler.) This perspicuous presentation makes possible that understanding which consists just in the fact that we “see the connections.” (cited in WDG 311)

Wittgenstein never wavered from this new methodological commitment, and an

essentially unmodified restatement of it appears in thePhilosophical Investigations: “A

perspicuous representation produces just that understanding which consists in ‘seeing

connexions’.... The concept of a perspicuous representation is of fundamental

significance for us. It earmarks the form of account we give, the way we look at things”

(PI §122). Wittgenstein’s notion of perspicuity appears to be directly indebted to

Spengler’s “physiognomic flair”: perspicuity characterizes a form of sight (“the way we

look at things”), and a corresponding form of description, which is analogous to

Spengler’s characterization of the way we perceive both faces and cultures (“without

any deliberate effort or system” and “far removed from cause and effect”). Just as

Spengler’s physiognomic or morphological analysis of cultures can reveal connections

that evade the grasp of casual explanation, Wittgenstein’s perspicuity characterizes a

form of sight and description that, as he puts it, “produces just that understanding which

consists in ‘seeing connexions’.” One might say that for both Wittgenstein and

Spengler, understanding and meaning are produced by, and as, a physiognomy.

Or to put it simply: “Meaning is a physiognomy” (PI §568). Seeing the

connections between the physiognomic methods of Spengler and Wittgenstein helps us

now to understand this remark, which appears parenthetically following a discussion of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. whether a certain rule (that “the kings are to be used for drawing lots before a game of

chess”) should be considered essential or inessential to the game. Using the language of

physiognomy, Wittgenstein concludes: “If I understandcharacter the of the game

aright—I might say—then this isn’t an essential part of it. ((Meaning is a

physiognomy.))”' (PI §568, emphasis added). The importance of this discussion lies in

its relevance to the related issues of meaning and rule-following: if the meanings within

a given language-game are determined solely by the following of rules, then there

should be no possible (or at least meaningful) recourse beyond those rules - indeed,

what would constitute such a beyond? There would be no sense, if this were the case,

in making a distinction between the essential or inessential rules of a given game, and

Wittgenstein gives voice to this view when he exclaims in the midst of this discussion:

“But, after all, the game is supposed to be defined by the rules!” (§567). But in propria

persona, he makes it clear that rules do not exhaust the parameters of a game: “The

game, one would like to say, has not only rules but also apoint” (PI §564, emphasis in

original). Wittgenstein’s introduction of the concept of a “point” suggests that meaning

is not completely determined by the totality of the rules of a game. The point of a game

cannot be gathered by reading off a complete list of its rules; it is something graspable

in addition to those rules (“not only rules butalso a point”). Yet there is no suggestion

that the point of a game is some kind of mental or spiritual entity added to the rales.

Indeed, to take the point of a game as something simply added to the totality of rales

would leave it indistinguishable from a rule at all, or would force us to call it some kind

of super-rule which controlled the sense of all the other rales of the game, allowing us

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to determine which of them were essential or not. Wittgenstein.5 s idea of a “point” is

clearly different in kind, and not simply in degree, from the concept of a rule; yet it is

difficult to know how further to specify what a “point” is. But it is not clear that

Wittgenstein means us to be able to. Thepoint of a game—and the idea of a “point”

here suggestively recalls Spengler’s concept of destiny—seems to be something one can

simply see, as Spengler said one could read a fate from a face. It appears that

Wittgenstein can assert that a rule is inessential to the game of chess simply because he

grasps the game’s “character” - in the form of its physiognomy. Strange as this may

sound, it seems that for Wittgenstein, meaningis a physiognomy, and understanding a

kind of physiognomic flair.

Strange indeed. There is an almost mystical quality to the emphasis on direct

seeing expressed in these physiognomically inflected remarks of Wittgenstein. Yet to

call his flair for physiognomy “mystical” would misleadingly give the impression that

what he regards as meaningful about any given physiognomy (whether of a game, a

word, or an actual face) is somehow hidden, transcendent or otherwise mysterious. This

could not be farther from the truth: for the radical innovation of Wittgenstein’s

physiognomy, as with the whole of his later philosophy (and we can now appreciate, I

think, how significant physiognomy is to the later writings), consists of his view that

meaning and mentality are not occult phenomena which accompany the physiognomy

of a game, a word, or a face, but are rather directly visible to anyone who cares to look.

On our ability to perceive mentality, or consciousness, in the faces of other people, he

writes:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Consciousness in another’s face. Look into someone else’s face, and see the consciousness in it, and a particularshade of consciousness. You see on it, in it, joy, indifference, interest, excitement, torpor and so on. The light in other people’s faces. Do you look into yourself in order to recognize the fury inhis face? It is there as clearly as in your own breast. (Z 220)

We need not refer, by analogy, to our own mental states (that is, look into ourselves) in

order to perceive the mentality of another (as claimed, for example, by Edmund

Husserl’s phenomenological theory of intersubjectivity in theCartesian Meditations);

nor, for that matter, is there any need to refer to some private psychological realm

within the other person, lying behind or beneath the expressions of his or her face.

No inner facts at all, whether in ourselves or in the other person, are required in

order to explain our ability to directlysee mentality in the appearance of another’s face.

Wittgenstein’s point is simply that such direct seeing is alwayspossibility a (an

everyday occurrence even), not that we can have epistemologicalcertainty with regard

to the mental states exhibited by the expressions of others - that, of course, would be

patently implausible. Without a doubt, the other person could be feigning or

suppressing their facial expressions; just as we may be deceived, or simply mistaken,

about their significance. Indeed, perhaps it is the very real possibility of these kinds of

deceptions and mistakes that gives rise to the common picture of human psychology

that postulates a division between the inner and the outer, the inner constituting a

privileged and essentially private domain accessible solely by subjective introspection.

Consider, for example, the following passage from Augustine’sConfessions which

imagines the mind, even of an infant, in precisely those terms:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Little by little I began to realize where I was and to want to make my wishes known to others, who might satisfy them. But this I could not do, because my wishes were inside me, while other people were outside, and they had no faculty which could penetrate my mind. So I would toss my arms and legs about and make noises, hoping that such few signs as I could make would show my meaning, though they were quite unlike what they were meant to mime.61

In this passage, the contents of Augustine’s mind, including his “wishes,” are quite

explicitly represented as fully formed even prior to the acquisition of language. Later,

words would be learned that could point to, and therefore publicly communicate, these

inner mental states, but the fundamental privacy and inwardness of these states is made

perfectly clear.

As is well known, however, Wittgenstein’s remarks on the idea of a “private

language” in theInvestigations show this Augustinian (and, of course, Cartesian and

later, Lockean) picture of mind to be conceptually incoherent (recall that the

Investigations opens with a critique of a passage about language learning from

Augustine’s Confessions). A private language, as Wittgenstein conceives of it, would

be one in which the words “refer to what can only be known to the person speaking; to

his immediate private sensations. So another person cannot understand the language”

(PI §243). Our ordinary talk of sensations do not fall under this definition of a private

language, for phrases such as “my arms itch,” or “my head throbs,” or “I feel pain in my

lower back” are perfectly understandable public expressions. A problem arises,

however, when these phrases are treated asreports that are distinct from the sensations

61 St. Augustine, Confessions (Harmondsworth: Penguin, 1961), p. 25.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that they are understood to describe. According to this view, the sensation itself (e.g.,

pain) is inner and absolutely private, and the outer public report describing it merely

associated with it by an introspective act of ostensive definition (i.e. THIS is what the

word “pain” refers to, and only I can know the sensation itself; correlatively, I can only

know the pain of others by analogy to THIS).

Questioning this way of regarding sensations, Wittgenstein asks, “how does a

human being learn the meaning of the names of sensations?—of the word ‘pain’ for

example” (PI §244). Countering the descriptivist model of sensation-language required

by Augustinian psychology, Wittgenstein proposes an expressivist one: “words are

connected with the primitive, the natural, expressions of the sensation and used in their

place. A child has hurt himself and he cries; and then adults talk to him and teach him

exclamations, and later, sentences. They teach the child new pain-behaviour.... [T]he

verbal expression of pain replaces crying and does not describe it” (PI §244). Contrary

to the Augustinian model, Wittgenstein makes no recourse to private objects of any

kind. Sensation-words do notdesignate private objects but rather express feelings in

the form of publicly shareable (and thus meaningful) verbal behavior. Wittgenstein

clearly finds this view acceptable, and the Augustinian one not. But simply presenting

an alternative model of sensation-language would do little to change the minds of those

who find it intuitively incontrovertible that private sensations are, at bottom, what

public sensation-words refer to. So Wittgenstein goes on to present a thought-

experiment in which he attempts to imagine an actual scenario in which words or terms

would be understood to designate private objects:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Let us imagine the following ease. I want to keep a diary about the recurrence of a certain sensation. To this end I associate it with the sign “S” and write this sign in a calendar for every day on which I have the sensation.—I first want to remark that a definition of the sign cannot be formulated.—But still I can give myself a kind of ostensive definition.—How? Can I point to the sensation? Not in the ordinary sense. But I speak, or write the sign down, and at the same time I concentrate my attention on the sensation—and so, as it were, point to it inwardly.—But what is this ceremony for? For that is all it seems to be! A definition surely serves to establish the meaning of a sign.—Well, that is done precisely by the concentrating of my attention; for in this way I impress on myself the connexion between the sign and the sensation. (PI §258)

At first, this description of a private language might seem unproblematic, though

Wittgenstein clearly has his doubts, as his jibe about a “ceremony” makes evident. It is

not, however, clear what, if anything, might be objectionable about the kind of

introspective act of ostensive definition that he imagines here (“S” = private sensation).

But, in fact, Wittgenstein finds this whole scenario incoherent:

But “I impress it on myself’ can only mean: this process brings it about that I remember the connexionright in the future. But in the present case I have no criterion of correctness. One would like to say: whatever is going to seem right to me is right. And that only means that here we can’t talk about ‘right’. (PI §258)

The problem is simply that it is conceptually incoherent to speak of the term “S” being

given any kind of definition at all in a private language, since there is no possibility of a

solitary speaker drawing a coherent distinction between correct and incorrect usages of

words that refer to private objects. In “the present case” of private ostensive definition,

anything would go, and thus no particular definition would be meaningful. It follows,

therefore, that such private objects cannot ground our understanding of the meaning of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sensation-words or, for that matter, the meanings of concepts which apply to mental

states of any kind (such as emotions, intentions, meaning or understanding). The point

is not that such private entities do not exist: any attempt to deny them would already

presuppose the meaningfulness of any talk about them, and would therefore fall prey to

the same critique. The point is simply that when we understand or use a psychological

concept (such as the “fury” Wittgenstein notes we can simplysee upon the face of

another), this understanding or use does not require recourse to some private inner fact,

whether in ourselves or in other people.

But if meaning and mentality are not explained by hiddeninner facts, this does

not mean that they are reducible to simplyouter ones either: Wittgenstein’s later

philosophy is neither a crude behaviorism nor a reductive physicalism. Indeed, as John

Koethe has argued (following Saul Kripke’s pioneering lead), Wittgenstein’s view is

that meaning and mentality are not reducible to facts of any kind atft, "7all. Certain verbal

and non-verbal behaviors (such as trembling, an accelerated pulse, or saying “I’m

afraid”) may serve as definitional elements (or criteria) of what we mean by the concept

fear, yet fear is not reducible to those, or any other, behaviors (and here we must resist

the temptation to say: well, that’s obvious, fear is THIS inside me!). According to

Koethe, criteria are “forms of behavior (among other things) that manifest or exhibit

62 See Koethe’s The Continuity o f Wittgenstein ’$ Thought, which has strongly influenced the interpretation of Wittgenstein in this essay. It is also due to Koethe’s suggestion, in the final chapter of his book, that Wittgenstein’s notion of direct seeing might profitably be compared to the philosophical theories of Dennett and Davidson that first led me to explore the connectionsI exploit that in the present chapter. John Koethe, The Continuity of Wittgenstein’s Thought (Ithaca: Cornell Univ. Press, 1996). References to this text will be given parenthetically with the abbreviation CWT. For Kripke’s reading of Wittgenstein, see: Saul Kripke, Wittgenstein on Rules and Private Language (Cambridge, MA: Harvard Univ. Press, 1982).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mental states,” but criteria are not equivalent to those mental states: fear-behavior is not

fear, though we only see fear manifested in the form of fear-behavior (CWT 110-111).

Mentality, in other words, is reducible to neither inner statesnor outer behavior. The

conceptual implications of this two-fold denial are startling, as is made evident by the

following exchange inZettel:

“But I do have a realfeeling of joy!” Yes, when you are glad you really are glad. And of course joy is not joyful behavior, nor yet a feeling around the comers of the mouth and the eyes. “But ‘joy’ surely designates an inward thing.” No. “Joy” designates nothing at all. Neither any inward nor any outward thing. (Z 487)

This striking conclusion about the non-factuality of mental states leaves us facing a

perceptual paradox: on the one hand, Wittgenstein argues that our words for mental

states (like joy and fury) designatenothing at all, while on the other hand, he insists that

mentality is immediately and directly visible in the face of another (“Do you look

within yourself, in order to recognize the fury inhis face? It is there as clearly as in your

own breast”). How are we to reconcile these apparently conflicting claims?

We can dissolve this conceptual problem if we turn our attention from the

question of what we see (the factuality of mind) to the more fruitful question of how we

see (the recognition of mentality). If, despite the non-factuality of mental states, we can

nonethelesssee mentality in another’s face, that is only because of the attitude (like

Dennett’s “intentional stance”) that we take towards the appearances and expressions of

mind in human forms of life. As Koethe puts it: “Thus to call a face, a body, or a form

of behavior a picture of mentality—or to say that we recognize mentality in it—is a way

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of characterizing our direct and naturalresponse or attitude toward it” (CWT 112;

emphasis in original). The attitude which discloses mentality is not ratiocinative in

nature: when we see that someone is joyful or enraged, this is not due to a process of

inference or interpretation. As Wittgenstein puts it in a passageZettel in that we have

already cited: “We do not see facial contortions and make inferences from them... to

joy, grief, boredom. We describe a face immediately as sad, radiant, bored” (Z 220).

Or as he puts it in the Investigations, “My attitude towards him is an attitude towards a

soul. I am not of the opinion that he has a soul” (PI p. 152). Consciousness, or

mentality, in other words, is simply and quite directlyrecognizable — even if the

conceptual account of such recognitional seeing will not necessarily seem so simple or

direct.

This does not mean that we do not occasionally need to consciously interpret or

make explicit inferences about someone’s appearance or behavior to ascertain their

mental state: perhaps a good friend’s disposition has suddenly become strange to us and

we do not understand the significance of her “stony” facial expression (is she

depressed? quietly seething? or on the verge of exhaustion?) - it would behoove us to

think carefully about what might have happened, or even what we may have

unknowingly done, to bring about this change. But lest we generalize such experiences

as this, and conclude that all instances of facial perception can be conceptually reduced

to conscious acts of interpretation, Wittgenstein offers the following wry remarks in

Zettel:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I interpret words; yes-—but do I also interpret looks? Do I interpret a facial expression as threatening or kind?—Thatmay happen. Suppose I said: “It is not enough to perceive the threatening face, I have to interpret it.”-—Someone whips out a knife at me and I say “I conceive that as a threat.” (Z 218)

Wittgenstein acknowledges that we may occasionally need to interpret the “look” of a

face, but as the negative example of the knife-wielding assailant makes clear, it would

be wrong to believe that interpretation is the way we regard faces in general. So while

he does not claim that direct seeing of mental states is the only form facial perception

takes, Wittgenstein holds that direct understanding of a mental state in a person’s face is

absolutely common (and also philosophically significant), and that, in fact, the attitude

that discloses mentality in the appearance of another’s face is more fundamental than

the ratiocinative one we occasionally need to employ. Even in those cases where we are

unsure as to what mental state a certain person’s face might be expressing at a given

moment, it makes little sense to say—unless the person is in an unusual condition, such

as a coma—that we do not at least recognize mentality as such (what would it mean to

sincerely think otherwise?). And when we do need to “think” about what a puzzling

face means, the interpretive reasoning we bring to it does not usually concern the spatial

organization of the facial features (e.g., does the contraction of his forehead

musculature indicate anger?), but naturally tends to focus on psychological questions of

a narrative nature (e.g., he looks like he might be angry... I wonder if so-and-so insulted

him again?).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Consciousness—-“the light in the face of another”—shines forth even if we are

uncertain as to what particular aspect of mind we are seeing. When looking at a face,

our regard does not choose between, on the one hand, a certain configuration of bone,

cartilage, muscle and skin, and on the other hand, an appearance of joy, for example;

but rather amongst such possible mental states as “joy, indifference, interest,

excitement, torpor and so on” (Z 220). The face is not perceptually constituted in parts

(as in an empiricist view of sense-perception), but rather the facefaces us all at once,

instantaneously, as a meaningful expression of mind — even when we cannot see the

whole of it at once (think of the deep significance we can glean from a sideways

glance). Wittgenstein found it conceptually significant that we can often grasp the

expression of a face while at the same time being unable to give any spatial description

of its features: “One may note an alteration in a face and describe it by saying that the

face assumed a harder expression - and yet not be able to describe the alteration in

spatial terms. This is enormously important” (WR 221).63 To Wittgenstein, this

absolutely ordinary fact about facial perception indicated the deep poverty of empiricist

models of visual perception: “One may also say: ‘He madethis face’ or ‘His face

altered like this’, imitating it - and again one can’t describe it in any other way. ((There

just are many more language-games than are dreamt of in the philosophy of Camap and

others.))” (WR 221). Those committed to sense-data theories of perception, for

example, cannot account for the obvious visibility of the “glance” in someone else’s

eyes:

63 Ludwig Wittgenstein,The Wittgenstein Reader , Ed. Anthony Kenny (Oxford: Blackwell, 1994). References to this text will be given parenthetically with the abbreviation WR.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. “If you only shake free from your physiological prejudices, you will find nothing queer about the fact that the glance of the eye can be seen too.” For I also say that I see the look that you cast at someone else. And if someone wanted to correct me and say that I don’t reallysee it, I should take that for pure stupidity. (Z 223)

The blindness of empiricism to the human gaze follows from its conception of the eye

as a passive organ of sensation, yet Wittgenstein notes that this theoretical assumption is

not tine to our everyday ways of regarding the powers of the eyes:

We do not see the human eye as a receiver, it appears not to let anything in, but to send something out. The ear receives; the eye looks. (It casts glances, it flashes, radiates, gleams.) One can terrify with one’s eyes, not with one’s ear or nose. When you see the eye you see something going out from it. You see the look in the eye. (Z 222)

But our ability to see sight itself cannot be explained by merely broadening the reach of

the senses. The “glance” is not an empirical fact like the color of an iris, or the dilation

of a pupil: “I should contradict anyone who told me I saw the glance ‘just the way’ I see

the shape and colour of the eye” (Z 223). But then what kind of sight is it (if not that of

“Carnap and others”) that cansee something like a flashing glance, the “hardness” of a

face, or the fury of a friend?

Wittgenstein found the question of seeing so intriguing and philosophically

significant that he devoted the longest section of Part II of theInvestigations to its

consideration: sec. xi, on seeing aspects and aspect-blindness. As we have already

noted, Wittgenstein associated the term “aspect” with Weininger’s physiognomic

writings, and, as this connection suggests, the questionfaces of was central to

Wittgenstein’s interest in aspects of all kinds (not only those of faces, but also cubes,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. duck-rabbits, and words as well). Indeed, one primary but unstated goal of

Wittgenstein’s treatment of aspect-seeing, I would argue, was to produce a

philosophical description of “seeing” that would account for our physiognomic ability

to recognize different aspects of mind in the faces of others. Sex In and Character ,

Weininger claimed that geniuses have particularly fluid and changeable facial

expressions: “Physiognomists, therefore, must not be surprised that men of genius, in

whose faces a new side of their minds is continually being revealed, are difficult to

classify” (SaC 65). Weininger found his theory confirmed when he compared a number

of portraits, executed at different times, of such figures as Goethe, Beethoven, Kant and

Schopenhauer. ‘The number of different aspects that the face of a man has assumed,”

he concluded, “may be taken almost as a physiognomical measure of his talent” (SaC

65). Wittgenstein took this section ofSex and Character to heart: we know, for

example, that he took Weininger’s distinction between talent and genius very seriously

(in one particularly tormented and disturbing entry in his notebooks, he declared that he

was “no more than talented”).64 We can imagine Wittgenstein carefully studying these

physiognomic remarks in Sex and Character and wondering: what does it mean to say

that one can see a “new side” or “aspect” of mind in the appearance of a face? I believe

that Wittgenstein’s discussion of aspects was essentially an attempt to answer this

pressing and fundamental question about physiognomy.65

64 Culture and Value, p. 19. See Ray Monk biography for Weiningerlan and psychological context of this remark, p. 316. 65 Even the fact that Weininger compared faces inportraits seems to have caught Wittgenstein’s attention: sec. xi contains a number of remarks on the way we regard portraits, such as: “I say: ‘We regard a portrait as a human being,’—but when we do we do so, and for how long?Always, if we see it at

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In the Investigations, he addresses the question of facial aspects directly in

§§536-537 (earlier versions of these sections, as we noted earlier, make explicit

reference to Weininger) by asking what it would mean to see a timid faceas

courageous:

I say: “I can think of this face (which gives an impression of timidity) as courageous too.” We do not mean by this that I can imagine someone with this face perhaps saving someone’s life (that, of course, is imaginable in connexion with any face). I am speaking rather of an aspect of the face itself. Nor do I mean that I can imagine that this man’s face might change so that, in the ordinary sense, it looked courageous; though I may very well mean that there is a quite definite way in which it can change into a courageous face. (PI §536)

Wittgenstein makes it clear that when he claims to be able to see a timid faceas

courageous, he is not making any empirical claims, either about what a person with

such a facial expression might be capable of doing, or about a possible change in the

spatial configuration of the face’s features. He insists that he is speaking of a change in

the way he can regard the face, which he associates with the idea of “an aspect of the

face itself.” Without any actual change occurring in the appearance of the face, he is

simply able to see it in a new light: “a quite definite way in which it can change into a

courageous face.” Aware, perhaps, that his point is rather unclear, he approaches the

same question from a different angle in the following section:

It is possible to say “I read timidity in this face” but at all events the timidity does not seem to be merely associated, outwardly connected, with the face; but fear is there, alive, in the features. If the features change slightly, we can speak of a corresponding change in the fear. If we were asked “Can you think of this face

all (and do not, say, see it as something else)? I might say yes to this, and that would determine the concept of regarding-as” (PI p. 175).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. as an expression, of courage too?”—we should, as it were, sot know how to lodge courage in these features. Then perhaps 1 say “I don’t know what it would mean for this to be a courageous face.” But what would an answer to such a question be like? Perhaps one says: “Yes, now I understand: the face as it were shews indifference to the outer world.” So we have somehow read courage into the face. Now once more, one might say, courage fits this face. But what fits what here? (PI §537)

In this passage, Wittgenstein elaborates on what might be involved in seeing a face

under a new aspect: in this instance, it would involve incorporating the timid face into a

new kind of psychological narrative, which would permit us to see the unchanged facial

expression in a different light. Wittgenstein seems satisfied that this account does

indeed explain one possible way of seeing a new aspect of the face (“Now... courage

fits this face”), yet his closing question suggests that he is a bit troubled by (or even

uncertain about) the potential implications of his explanation. His question—“Butwhat

fits what here?”—seems to warn us from taking the appearance of the face and the

meaning of its expression as distinguishable entities: as a subjective reading or

interpretationfitting into the objective appearance of the face-itself. His fear that we

may misinterpret his account this way seems to be the point behind the opening of the

passage, which prepares us for the aspect-conversion by insisting on the

indistinguishability of the face from its expression of timidity. He admits that it is

meaningful to say “I read timidity into this face,” but insists that we should not regard

the feeling as “merely associated” with the face. The mental state is not “outwardly

connected” to the face’s appearance, but is rather “there, alive,in the features.” Hence

the point of his question: “But what fits what here?” Having claimed that we can

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. incorporate the face into a new psychological narrative (timidity becoming visible as

courage), Wittgenstein seems to worry that this suggests that aspects are arbitrary,

subjective associations we cm fit into our visual impression of a face’s appearance.

While it is clear that he thinks it incorrect to think of aspects as subjective

interpretations of visual impressions, he breaks the discussion off at this point and does

not return to the question of aspects until he treats it at length in section xi. Despite the

fact that he also examines such diverse phenomena as schematic cubes, triangles and

duck-rabbit illustrations, the discussion of seeing aspects and aspect-blindness in

section xi clearly extends and elaborates upon the question of facial aspects that

concerned Wittgenstein in §§536-537 (the only other passages inInvestigations the that

discuss the subject of aspects at all). In the opening lines of section xi, Wittgenstein

obviously still has the human face in mind as he begins his discussion of aspects by

distinguishing between two different kinds of seeing:

Two uses of the word “see”. The one: “What do you see there?”—“I seethis” (and then a description, a drawing, a copy). The other: “I see a likeness between these two faces”—let the man I tell this to be seeing the faces as clearly as I do myself. The importance of this is the difference of category between the two ‘objects’ of sight. The one man might make an accurate drawing of the two faces, and the other notice in the drawing the likeness which the former did not see. (PI p. 165)

Wittgenstein finds it significant that one person can notice something about two faces (a

likeness) that another person, who also sees the very same faces, might not. He is

careful to establish that this difference wouldnot be due to any differences in the visual

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. facts perceptually available to the two people (e.g., differences in angle, lighting,

disposition of features, etc.): in fact, he goes so far as to say that the likeness can be

perceived by the second personin an “accurate drawing” made by the first. The two

people, in other words, see the very same faces - only they see them in very different

ways. The curious visual experience of the second person captures what Wittgenstein

means by aspect-seeing: “I contemplate a face, and then suddenly notice its likeness to

another. I see that it has not changed; and yet I see it differently. I call this experience

‘noticing an aspect”’ (PI p. 165). An aspect, therefore, is evidently not equivalent to an

exhibition of mentality, though, as with the courage in §536, this is clearly one of its

possible, and I would suggest primary, meanings. Rather, it more generally signifies

something visible in an object—certainly in faces, and later, in a variety of other objects

besides—that exceeds (like a glance) what is “empirically” or “factually” available to

sight: something, to recall Wittgenstein’s Shakespearean quip, not dreamt of in the

Horatian philosophies of “Carnap and others.”

Since the two people in Wittgenstein’s example see exactly the same visual

facts, we might think—as he appears to have been worried that we might think of the

courageous expression in §537—that the difference in visual experience is due to an

interpretation introduced by the second person. This explanation would certainly be

consistent with empiricist and phenomenalist models of sense-perception that posit a

private visual impression (or “inner picture,” as Wittgenstein calls it) constituted from

the data of sensation. According to this view, since an aspect is not anything actually

attributable to the object, it must be an interpretation originating in the subject.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wittgenstein, however, finds the “inner picture” model of visual perception incoherent.

To illustrate one problem with it, he introduces the following schematic illustration:

Fig. 2

He asks us to imagine that we find this illustration in several places throughout a text

book, each time accompanied by a different textual gloss instructing us how to regard

the figure: “here a glass cube, there an inverted open box, there a wire frame of that

shape, there three boards forming a solid angle. Each time the text supplies the

interpretation of the illustration” (PI p. 165). He notes that as we read each set of

interpretive instructions, we not only can think of the figure that way, but can actually

see the figure differently each time: “we can alsosee the illustration now as one thing

now as another.—So we interpret it, andsee it as we interpret it.” The similarities to

the discussion in §537 of the timid and courageous face are striking: there he came up

with a new psychological interpretation that allowed him tosee the timid face

differently, even though, as with the schematic illustration seen in different ways here,

nothing “factual” about the spatial configuration of the face had changed.

And as with the conclusion of §537, the question arises as to how we should

understand the new visual experiences prompted by the various interpretive glosses on

the schematic illustration: in other words,what fits what here? For instance, if we see

the figure as an “inverted open box,” it might make sense to think of our visual

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. experience as one that interprets what we see in that particular way. In other words, we

have a direct visual impression of the schematic diagram which, when provided with the

appropriate instructions, we can interpretas an upside down box. As Wittgenstein puts

it, this would be to characterize the visual experience of the inverted open box as an

indirect one:

Here perhaps we should like to reply: The description of what is got immediately, i.e. of the visual experience, by means of an interpretation—is an indirect description. “I see the figure as a box” means: I have a particular visual experience which I have found that I always have when I interpret the figure as a box or when I look at a box. (PI p. 165)

In other words, the interpretation must be something which indirectly modifies our

direct visual impression of the schematic illustration. But Wittgenstein goes on to

question the use of the word “indirect” in this explanation. He notes: “But if it meant

this I ought to know it. I ought to be able to refer to the experience directly, and not

only indirectly. (As I can speak of red without calling it the colour of blood.)” (PI p.

165). In other words, the problem with calling our visual experience of the figure (as an

inverted open box) anindirect one implies that there is some more direct way of

describing that particular visual experience as well. But no more direct alternative is

available, as it would be if we were to say that a red object is “the colour of blood,”

since we could also simply call the object red. The box-ness of the figure that we see (if

we see it as a box) is simply there, not something we have added to a more direct visual

experience of an arrangement of line-segments. Of course, one might suffer from what

Wittgenstein calls “aspect-blindness,” and be unable to see the box-aspect (i.e. not be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. able to see it as a box), in which case one would have to infer that such an arrangement

of lines, according to certain illustrative conventions, canunderstood be as a box. But

this would be a different relationship to the box altogether, ratiocinative rather than

visual: for the aspect-blind, the words “indirect” and “interpretation” would be

appropriate to describe how they regard the figure as a box. But if we actuallysee the

figure as an inverted open box, what more direct way is there to describe our visual

experience?66

The implications of this example clearly apply to the discussion of the aspect of

courage in §537. In response to Wittgenstein’s closing question there—“Butwhat fits

what here?”—we can now say with certainty that it isnot an interpretation that fits onto

a visual impression: in fact, there are not two “what’s” involved at all. Wittgenstein’s

cautionary question is simply rhetorical in nature. If we can discern courage in a timid

face, it is no more “outwardly associated” with the features than was the original

expression of timidity. Courage simplyis an aspect of the face. But this does not mean

that we can willfully see anything anywhere we like: “I can see the schematic cube as a

box;—but can I also see it now as a paper, now as a tin, box?—What ought I to say, if

someone assured mehe could?—I can set a limit to the concept here” (PI p. 177). That

Wittgenstein made it clear that it was difficult to see the timid face as courageous in

§537—“we should, as it were, not know how to lodge courage in these features”—was

66 Interestingly, the research of Uta Frith and her colleagues has demonstrated that children with autism— who in many ways bear a striking resemblance to the aspect-blind in Wittgenstein’s later writings— show an Impairment in their ability to perceive visualgestalts. Instead of seeing cohesive wholes, they seem particularly sensitive to fragments and parts: autistics, that is, seem indifferent to contexts and the meanings of wholes. Cf. AEE, ch. 9: “A Fragmented World.”

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. perhaps intended to demonstrate that aspect-seeing, whilebeing “subject tothe will” is

not necessarily easily, or even wholly, so (PI p. 182). An aspect of mind is not in the

mind of the beholder, but is, as Wittgenstein put it quite clearly, “

But, of course, we do not usually go around asking ourselves if we can manage

to see this or that person’s face as joyful or angry. Mentality strikes us, faces us

directly. Wittgenstein’s excursus on the schematic box is invaluable in helping to

clarify the concepts involved in our understanding of what it means to see somethingas

something: the figure as a box or the featuresas courageous. But because it focuses on

experiences of aspect-seeing that are triggered by volition, it does not fully account for

our everyday recognition of aspects of mind in the faces of others, a kind of recognition

that is often characterized by an absolute absence of volition. “He must be in pain”...

“she looks relaxed”... “he seems anxious”: such common expressions are often

immediate responses to the way people look. To help us understand this type of

everyday form of non-volitional recognition, Wittgenstein discusses a variety of

phenomena that are related to what he calls aspect-dawning. To explain what he means

by this, he introduces Joseph Jastrow’s famous illustration of a duck-rabbit:

Fig. 3

He notes that “The picture might have been shewn me, and I never have seen anything

but a rabbit in it” (PI p. 166). Such a person would have continuously seen the rabbit-

aspect, and would be completely unaware that there was also a duck-aspect in the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. figure. If presented with the illustration, and asked what he saw, he would naturally

respond with something like, “I see a picture of a rabbit.” It would make no sense for

him to say, “I see this picture as a rabbit,” since what other possibility would exist?

But if this illustration were put amongst a number of unambiguous Images of

ducks, this person would likely recognize (that is, experience the “dawning” of) the

duck-aspect of the figure, and it is easy to Imagine some type of exclamation

accompanying this experience: “I saw it quite differently, I should never have

recognized it!” (PI p. 167). From this moment on, it would make sense for the person to

say: “I see this picture as a duck” or “as a rabbit.” Seen one way, and then another, the

duck-rabbit figure appears completely different, though it is just as clear that the figure

has not changed at all: “The expression of a change of aspect is the expressionnew of a

perception and at the same time of the perception’s being unchanged” (PI p. 167).

Wittgenstein asks: “But what is different: my impression? my point of view?” He

cautions against regarding the changed perception as the result of an alteration in the

“organization” of some inner visual impression: in other words, against theorizing that

the lines are subjectively re-organized now as a rabbit-picture, now as a duck-picture.

This would entail thinking of “organization” as a property of the visual impression, on

the same level as colour and shape. But “If you put ‘organization’ of a visual

impression on a level with colours and shapes, you are proceeding from the idea of the

visual impression as an inner object. Of course this makes this object into a chimera; a

queerly shifting construction. For the similarity to a picture is now impaired” (PI p.

168). If one thinks of the inner picture as being organized differently when the figure is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. seen as a rabbit or as a duck, then while this might explain the change in how we regard

what the figure represents, it would conflict with the presupposition that the inner

picture is a reflection of the outer picture in the first place, rendering the entire “inner

picture” theory of perception conceptually incoherent. According to the explanation by

re-organization, the inner picture would become a “queerly shifting construction” - no

longer an innerpicture at all. Moreover, if one asked two people who saw the figure

differently (one as a duck, one as a rabbit) to make copies of what they saw, the two

copies would be identical. One cannot point to anything factual in the figure that

accounts for it now appearing as a duck or as a rabbit: “If I saw the duck-rabbit as a

rabbit, then I saw: these shapes and colours (I give them in detail)—and I saw besides

something like this: and here I point to a number of different pictures of rabbits” (PI p.

168). The rabbit-aspect is not reducible to a factual property of the figure on a par with

shape and colour: “what I perceive in the dawning of an aspect is not a property of the

object, but an internal relation between it and other objects” (PI p. 180). That is why

Wittgenstein claims that ‘“Seeing as....’ is not a part of perception. And for that reason

it is like seeing and again not like” (PI p. 168).

Because visual experiences of aspects are not wholly perceptual in nature—as

sense-perception has been understood by empiricism and phenomenalism, at least—

verbal descriptions of aspects should not be understood as mere perceptual reports.

Wittgenstein claims that they should rather be regarded asexpressions of visual

experience (like cries of recognition):

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I look at an animal and am asked: “What do yon see?” I answer: “A rabbit”. —I see a landscape; suddenly a rabbit runs past. I exclaim “A rabbit!” Both things, both the report and the exclamation, are expressions of perception and of visual experience. But the exclamation is so in a different sense from the report: it is forced from us.—It is related to the experience as a cry is to pain. (PI p. 168)

It is not that the exclamation (and one can imagine a silent one to oneself) is without

conceptual content: after all, it too is a perceptual description of what is seen. What

differentiates an exclamation from a report, rather, is that the conceptual content of an

exclamation is not the result of a process of ratiocination, though one can still call it a

kind of “thinking”: “If you are looking at the object, you need not think of it; but if you

are having the visual experience expressed by the exclamation, you are thinkingalso of

what you see” (PI p. 168). One directly sees “arabbit /” and the exclamation expresses

the thought infolded into this visual experience, which Wittgenstein therefore

characterizes as a strange amalgam of sight and reason: “Hence the flashing of an aspect

on us seems half visual experience, half thought” (PI p. 168). Or as he puts it

elsewhere: “‘The echo of a thought in sight’—one would like to say” (PI p. 181).

Of course, this is a very strange idea: half sight, half thought? One can imagine

someone dismissing this idea as logically absurd. And indeed, one cannot be led to

accept or even see Wittgenstein’s point simply by means of logical argumentation.

Wittgenstein is not claiming that the concept of “seeing” that he is discussing is

theoretically deduciblea priori. Rather, he is simply describing an everyday kind of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. visual experience with which we are all perfectly familiar, but before which empiricist

theories of perception are conceptually blind. For example:

Someone suddenly sees an appearance which he does not recognize (it may be a familiar object, but in an unusual position or lighting); the lack of recognition perhaps lasts only a few7 seconds. Is it correct to say he has a different visual experience from someone who knew the object at once? (PI p. 163)

One might answer yesand no: no, for each person in this thought-experiment sees

precisely the same visual data (that is, they both can be understood to have identical

sensory impressions); butyes as well, for one person at first seesonly the visual data

while the other person immediately recognizes (like the running rabbit above) what the

object is. According to sense-data perceptual theories, however, the visual experience

of each person would be identical, one of them merely taking a bit longer to locate the

appropriate concept under which to place the visual impression. And perhaps the sense-

data theory is correct, for it does manage to account for the basic facts of the situation.

After all, “might not someone be able to describe an unfamiliar shape that appeared

before him just as accurately as I, to whom it is familiar?” (PI p. 168). For if both

persons experienced the same visual impression, as presupposed by the sense-data view,

then there should be no difference in the way each could describe the familiar or

unfamiliar object, for each would haveseen the same thing, only understanding it

differently. But, as Wittgenstein retorts, “Of course it will not generally be so. And his

description will run quite differently. (I say, for example, “The animal had long ears”—

he: “There were two long appendages”, and then he draws them.)” (PI p. 168).

Recognized, the shape is seenas an animal, and everything comes into view differently

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (those are not “appendages” butears). But seen without recognition, it is merely a

shape. The visual experienceitself is completely different depending on whether

recognition is involved or not: whether or not, in other words, one experiences “the

echo of a thought in sight,” however strange this may sound.

One implication of Wittgenstein’s notion of aspect-seeing is that, though the

experiencing of aspects is not entirely conceptual in nature, a mastery of certain

concepts is necessary to have certain aspect-related experiences. Without a mastery of

the concepts of “duck” and “rabbit” (which would include familiarity with their shapes),

one could not experience the dawning of their aspects in the duck-rabbit figure: “You

only ‘see the duck and rabbit aspects’ if you are already conversant with the shapes of

those two animals” (PI p. 177). Wittgenstein pursues this line of thought by discussing

a kind of aspect-experience that he calls “aspects of organization,” in which “when the

aspect changes parts of the picture go together which before did not” (PI p. 177):

In the triangle I can see nowthis as apex,that as base—now this as apex,that as base.—Clearly the words “Now I am seeingthis as the apex” cannot so far mean anything to a learner who has only just met the concepts of apex, base, and so on.—But I do not mean this as an empirical proposition. “Now he’s seeing it like this”, “now like that” would only be said of someonecapable of making certain applications of the figure quite freely. The substratum of this experience is the mastery of a technique. (PI p. 178)

Wittgenstein acknowledges that there is something quite odd about this claim, for he is

insisting that various forms of purely visual experience are unavailable to those who

have not adequately mastered certain conceptual “techniques”: “how queer for this to be

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the logical condition of someone’s having such-and-such experiencean ! After all, you

don’t say that one only ‘has toothache’ if one is capable of doing such-and-such” (PI p.

178). The point is not that one would simply be unable to describe the duck-rabbit

figure as a duck if one did not know the concept (including the shape) of a “duck,” but

rather that one would not even be able tosee it as a duck. A person who was clumsy

with, or ignorant of, the concept of a “duck” would have access to one fewer visual

experiences, not merely one fewer descriptive terms.

In other words, aspect-seeing revealsseeing to have its own kind of intelligence,

which therefore requires, as all forms of intelligence do, enculturation and education:

“For how could I see that this posture was hesitant before I knew that it was a posture

and not the anatomy of the animal?” (PI p. 178). Unless we have somehow become

intimately familiar with the concept of animal postures, we would be unable to directly

see a particular disposition of an animal’s bodyas this or that psychological state. If, on

the other hand, we happen to have been raised by and among humans who consistently

related to a particular kind of animal—dogs, for example—as relatively complex

psychological beings, we might very well be able to see hesitation manifested in the

mere curve of a dog’s spine. But for those who have not experienced such enculturation

into the psychic lives of dogs, an array of psychological concepts having to do with

canine postures (such as hesitation, aggression, submission and so on) would literally

not be visible: they would be blind to those aspects of a dog’s mental life - or perhaps it

would be more accurate to say that for them those psychological states would simply

not exist in a dog. Hence, the way humans perceive dogs varies widely from culture to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. culture: some societies treating them with the kind of care usually reserved for humans

(that is, “humanely”); others treating them as little more than a nuisance, sometimes

even as food.

The conceptual analogy with the psychological states visible in the human face

is obvious, and Wittgenstein does not neglect to consider it (again returning to the

example of §§536-537): “Might I n o t... have a purely visual concept... of a timid

face?” (PI p. 178).

We react to the visual impression differently from someone who does not recognize it as timid (in thefull sense of the word).— But I do not want to say here that we feel this reaction in our muscles and joints and that this is the ‘sensing’.—No, what we have here is a modified concept ofsensation. One might say of someone that he was blind to theexpression of a face. Would his eyesight on that account be defective? (PI p. 179)

Recognizing the aspect of timidity in a face is something that one who has mastered the

concept of timidity can do without a thought, so to speak. We can have a “purely visual

concept... of a timid face”: wholly visual, yet conceptually rich. But for one who has

not mastered the concept of timidity, a timid aspect would not be directly recognizable

upon the face of another, and this blindness would be made evident by the fact that he

would “react to the visual impression differently”: perhaps, for example, showing

indifference where another might have shown concern. The way a person responds, or

does not, to an aspect of mind in the face of anothershows whether that person can

experience (thatis, fully knows the concept for) that aspect of mind at all. Following

this line of thought, it is clear that an important ethical function of cultural practices

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. such as art and literature is the conceptual training (that is, sensitization) of humans to

recognize more and finer shades of mental experience, both in ourselves and others,

though it is, of course, always a possibility that such “sensitization” could be perverted

into a self-regarding and decadent end-in-itself. Seen as a kind of conceptual training,

art would not be understood as a form of “expression” (bringing out what is latent and

within) but rather as a kind of “cultivation” that plants the seeds of new intersubjective

cognitive possibilities. We should also say that those who are blind to a particular

expression on a face are not suffering from some type of physiological defect in their

eyesight, but rather from a kind of partial or selective mmd-blindness, which we can

always hope might be corrected by conceptual familiarization with the “unseen” state of

mind. However, we also now know that such sight is tragically beyond the reach of

those who suffer from severe autism, a biological form of full-blown mind-blindness.

As we have seen, research into the condition of autism confirms the intersubjective

implications of Wittgenstein’s theory of aspect seeing: seeing states of mind in other

h u m an s, like the ability to see a particular side of a triangle as the base, is the result of

the “mastery of a technique” which one can sadly lack.

The “modified concept of sensation” Wittgenstein uses to describe aspect-seeing

bears a striking similarity to the kind of direct seeing—or Spenglerian “physiognomic

flair”—that we saw implied by the phrase, “Meaning is a physiognomy” (PI §568). In

fact, Wittgenstein confirms the association we have already suggested between aspects

and physiognomy when he remarks that the experience of seeing a face might be

“explanation enough” for the concept of an aspect: “The aspect presents a physiognomy

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. [Physiognomic] which then passes away. It is almost as if there were a face there which

at first I imitate , and then accept without imitating it.—And isn’t this really explanation

enough?—But isn’t it too much?” (PI p. 179). Perhaps we see revealed in these

ambivalent remarks the underlying physiognomic motivation for the whole

investigation of aspects in the first place. The suggestion about “imitation” here recalls

the remark he made about facial perception in the context of his jibe about the

perceptual theories of “Carnap and others”: “One may also say: ‘He madethis face’ or

‘His face altered like this\ imitating it - and again one can’t describe it any other way”

(WR 221). Aspects, like the facial expressions in this remark, are clearlyvisible yet

impossible to describe in factual terms: an aspect isthere to be seen, but what is there?

Nothing, and yet everything that makes us “human.” Faces, and the aspects that play

across their features, inspired Wittgenstein to radically reorient our understanding of

seeing in general, turning us away from the question ofwhat we see (factuality) to that

of how we see (recognition). We could say that his entire investigation of aspects

emerged from his sense that no available theory of perception could help him answer

one basic question about the foundations of physiognomy: what does it mean to say that

one can simply recognize mind in the appearance of a person’s face?

Wittgenstein’s discussion of aspect-seeing, however, has conceptual

implications that go far beyond its initial focus on the human face and the sense of

sight. For example, in his discussion of the timid and courageous face in §536,

Wittgenstein compares the experience of aspect-conversion with that of tonal

modulation: “The reinterpretation of a facial expression can be compared to the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reinterpretation of a chord in music, when we hear it as a modulation first into this, then

into that key.” Tonal modulation is not something factually inhering in the notes of a

given chord, but neither is it something merely subjectively apprehended in the mind of

the listener, as when, to take a simple example, we can hear the notes C, E and G played

together as the tonic of C Major, and then as the sub-dominant of the key of G. Just as

courage was called “an aspect of the face itself,” tonality can be thought of as an

auditory aspect of the musical notes (PI §536). Later, in the context of his discussion of

the “purely visual concept... of a timid face” in section xi, Wittgenstein again

associates aspects with various experiences of musical perception:

Think of this too: I can only see, not hear, red and green,— but sadness I can hear as much as I can see it. Think of the expression “I heard a plaintive melody”. And now the question is: “Does he hear the plaint?” And if I reply: “No, he doesn’t hear it, he merely has a sense of it”—where does that get us? One cannot mention a sense- organ for this ‘sense’. (PI p. 178)

Wittgenstein’s remarks here make clear that though aspects are first discussed in the

context of visual perception, aspect-like experiences are not limited to the sense of

sight. Wittgenstein even suggests that “Aspect-blindness will akinbe to the lack of a

‘musical ear’” (PI p. 182). There is, after ail, no specific “sense-organ” associated with

aspect-experiences, and, in any case, aspects are not attributable to the factual contents

of any sensory manifold or impression. So it follows that they can be experienced in

any field of human life, not simply the visual, but the auditory too, and even the purely

conceptual as well, as suggested by the example of the direct seeing of the

“physiognomy” of the game of chess in §568.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. According to Wittgenstein, however, the most important implications of his

discussion of aspect-seeing are related to the concept of “experiencing the meaning of a

word,” for Wittgenstein thinks of words as he thinks of faces: “The familiar

physiognomy [Gesicht] of a word, the feeling that it has taken up its meaning into itself,

that it is an actual likeness of its meaning” (PI pp. 182, 186). A word, that is, exhibits

meaning as a face exhibits mind: while we may occasionally need to interpret a word,

just as we may sometimes need to think about the “look” of a face, the ordinary way of

experiencing the meaning of a wordrecognitional, is characterized by immediacy and

an absence of ratiocination. Wittgenstein acknowledges that “there could be human

beings to whom all this was alien” (PI p. 186). For them, a word’s meaning would

seem arbitrarily connected to its appearance, and—as with those who cannotsee

timidity in a face—this meaning-blindness would be shown by the way they behaved

towards words: “They would not have an attachment to their words” (PI p. 186).

Language, for such humans, would be like a code “in whose use the ‘soul’ of the words

played no part. In which, for example, we had no objection to replacing one word by

another arbitrary one of our own invention” (PI §530).

But, of course, such a picture of language does not capture our familiar

relationship to words. To illustrate his point, Wittgenstein asks us to imagine a

hypothetical case of word-substitution: “Suppose I had agreed on a code with someone;

‘tower’ means bank. I tell him ‘Now go to the tower’—he understands me and acts

accordingly, but he feels the word ‘tower’ to be strange in this use, it has not yet ‘taken

on’ the meaning” (PI p. 182). Wittgenstein is certainly not claiming that words have a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. necessary or a priori connection to their meanings. That is, he is not talking about some

kind of Adamic tongue (though Ms idea that we see words in the way we see faces does

happen to explain why humans would believe in the existence of a divine language in

which words and objects arenecessarily connected, as a face is to a mind). On the

contrary, the meanings of words are still to be understood as conventionally determined

by their use, and he even implies that the word “tower” in this example could eventually

(if “not yet”) take on its new meaning. Nor is he saying that the arbitrarily chosen word

in this scenario would be incomprehensible: the person in this example “understands ...

and acts accordingly.” The point, rather, is that linguistic theories that conceive of

words as arbitrary signs (like that of Saussure) are unable to adequately account for the

familiar ways in which we ordinarily use and understand (that is, relate to) words. The

meaning of a word is not merely associated or “outwardly connected” with its

appearance but is anaspect of the face (or appearance) of the word itself.

And just as we can see a variety of psychological states in a face, a word can be

understood in a number of different ways, or under different aspects. For example,

“You can say the word ‘March’ to yourself and mean it at one time as an imperative, at

another as the name of a month” (PI p. 184). A person who was blind to the “faciality”

of the appearance of a word would not understand what it feels like to take a word now

one way, now another: “What would you be missing, for instance, if you did not

understand the request to pronounce the world ‘till’ and to mean it as a verb” (PI p.

182). But because we are able to say “till” and mean it as a verb, and also as a noun,

and even as a preposition, we know that we are not blind to the meaning-aspects of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. words. We are already physiognomists of language: we naturally see words as we see

faces, along with the affective implications face-to-face relations often imply (the

aspect-blind, on the other hand, would “not haveattachment an to their words” [PI p.

186]). Put plainly, we care about words. Wittgenstein, therefore, finds it revealing that

we often obsess about finding the “right” word for a particular occasion:

How do I find the ‘right’ word? How do I choose among words? Without doubt it is sometimes as if I were comparing them by fine differences of smell:That is too..., that is too...,— this is the right one.—But I do not always have to make judgments, give explanations; often I might only say: “It simply isn’t right yet”. I am dissatisfied, I go on looking. At last a word comes:“That’s it!” Sometimes I can say why. (PI p. 186)

It is not unusual for us to care deeply about finding the right word for a particular use.

We might struggle to find it, but this does not mean that we necessarily cogitate in some

rational manner (i.e. “make judgments”), as though consciously processing the options

available in an internal dictionary or lexical memory-bank. We may not be able to infer

what the right word will be; but we can often recognize that we have found it. We may

not even be able to say how we found the right word, or even precisely why it is right.

We just see or hear that it fits.

Who has not had linguistic experiences like these? Yet this discloses a

dimension of our ordinary relationship to words that cannot be accounted for by

traditional philosophies of language. On the other hand, such experiences of looking

for and finding words are quite understandable if, along with Wittgenstein, we think of

words as having faces - and conceive of linguistic understanding as a form of

physiognomy. This, of course, does not mean that wordsliterally have faces: that is,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. have appearances which are somehow perceptually analogous to the features of human

faces. What would a claim like that mean? And where would it get us anyway? After

all, Wittgenstein has already established that if we see mentality exhibited by a face,

that experience is not attributable to anything factual about the face. If we see mind in a

face, it is not because mind is somehow literallythere, but because we take a

recognitional attitude or stance towards the appearance of the face’s features. So to say

that a word has a face is not to say that it somehow looks like a face, which would

establish nothing, but rather to say that we approach words in the recognitional way we

approach faces.

And, of course, Wittgenstein can understand words by analogy to faces only

because he already regards faces physiognomically. For him, the human face is the

paradigmatic form in which mind and meaning are directly recognizable: meaningis a

physiognomy. If references to physiognomy in the later writings are often meant in a

figurative or analogical sense (we see words or gamesas we see human faces), this

figurative usage is clearly and quite explicitly tied to the primary senses of

“physiognomy” as the human face itself and the art involved in its interpretation.

Wittgenstein’s later philosophical outlook, in other words, is profoundly

anthropomorphic: not, however, because he projects the human form beyond the class

of human beings, but rather because he implies that humanity itself is the outcome of a

kind of anthropomorphic projection. The recognition of mentality in the face of another

person is not grounded upon some deep fact of the matter about that person’s mind, but

is simply the result of the intentional stance we naturally adopt towards him or her: “My

100

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. attitude towards him is an attitude towards a soul,” as he put it (PI p. 152). Though

their respective idioms are quite different, Wittgenstein’s explanation of the way we see

aspects of mind In the human face is not fundamentally unlike Quine’s notion of the

“essentially dramatic idiom” that we use to attribute propositional attitudes to

intentional entities. In each case, it is explicitly denied that attributions of mentality are

grounded upon the existence of private mental entities. If we see mentality all around

us, that is not because there are such things as minds: as Davidson rather dryly states the

point, “There are no such things as minds, but people have mental properties” (CPM

231).

We might say that Wittgenstein encourages us to take others at face value: that

is another way of saying that he teaches us to properly value the face. Physiognomy is

possible and necessary not because the outward face reliably mirrors the inward soul (as

the physiognomic tradition, including Weininger, mistakenly believed), but rather

because there is nothing behind the face to which we can turn. After all, appearances

and behavior (including, of course, linguistic testimony) are all we have to go on, and

there are no deeper facts that will settle residual uncertainties about competing

mentalistic attributions. Thus, it might be possible to see a given face as “timid”or

“courageous”: if so, how is one to decide the matter? Simply by deciding the matter, as

we naturally do in any case. As Dennett puts it in a passage that we have already noted,

it is always possible (though unlikely) that “rival intentional interpretations” might arise

“that did equally well at rationalizing the history of behavior of an entity. Quine’s

thesis of the indeterminacy of radical translation carries all the way in, as the thesis of

101

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the indeterminacy of radical interpretation of mental states and processes” (CPM 239).

Wittgenstein’s discussion of the timid and courageous face in Investigationsthe

suggests that he would not have found Dennett’s point difficult to understand. But if,

following Wittgenstein, we detach the face from the inner mind that appears to animate

it, are we not left with a mere mask? What then becomes of ail our notions about the

person?

• Dramatic Idioms: Personae and the History of Personhood

Considering Quine’s talk of “dramatic idioms,” it is curious, but probably not

accidental, that the modem word person derives from the Latin wordpersona, which at

first had an exclusively theatrical meaning: it referred to a mask used by an actor, but

could also designate the one who performed or the part acted (asdramatis in personae,

as we still say). fn In turn, the Latin wordpersona, by way of the Etruscan term phersu

(for mask), comes from the Greek word prosopon, which could also mean a mask, a

dramatic part, or a face. The metaphoric connections to faces, masks and theater that

are implicit in the history of the word person “can still haunt us,” Raymond Williams

observes in his etymological discussion of the English word “personality” (in

Keywords) 68 But the attenuation implied by Williams’ use of the word “haunt” is

67 Etymology is given in A. David Napier,Masks, Transformation, ana Paradox (Berkeley: Univ. of California Press, 1986), p. 8 and p. 20. References to this text will be given parenthetically with the abbreviation MTP. 68 Raymond Williams,Keywords: A Vocabulary of Culture and Society, rev. ed. (New York: Oxford Univ. Press, 1983), pp. 232-235.

102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. telling: while the theatrical associations implicit in its etymological history still have

some bearing on our understanding of the word “person,” such associations constitute a

rather pale and ghostly semantic aura. Moreover, the theatrical associations, while

certainly adding some color to a rather abstract legal and philosophical category, are

usually invoked only in a pejorative sense: to suggest that a person is playing a part,

wearing a mask, or fulfilling a role are not generally considered forms of high praise.

This has a great deal to do with the ambivalence about theatricality that lies deep in the

culture of the West, an attitude that can be traced to the rise of early Christianity. It is

the early Church that decisively purged the Latin wordpersona of its primarily dramatic

meanings and replaced them with the qualities of depth, interiority and subjectivity that

we associate with the modem self.

From the earliest days of the Church, the persona was a fundamental category in

Christian theology, referring not only to individual believers but also to the divine

persons of the Trinity. At the same time, however, the persistent theatrical associations

of the Latin wordpersona were a source of great tension in early Christian thought.

Anti-theatricality in general was vigorous and widespread among early Christian

writers, who believed that the classical theater was radically incompatible with a

Christian worldview. As Charles Garten puts it:

the continuing intent to glorify the god, the fundamentally religious character of theatrical events—that is, in Christian eyes, their fundamentally pagan character—is emphasized and reemphasized in patristic writings down to the end of antiquity....

103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A man cannot be a Christian and remain an actor, because he is serving pagan gods, and the audience are his abettors.69

In his anthropological study of masks, David Napier persuasively argues that profound

metaphysical differences motivated early Christian anti-theatrical polemics (rather than

mere disdain for the immorality of the theater, for example): “only a direct

epistemological incompatibility can account for the Church’s persistent criticism of the

stage” (MTP 11). Napier suggests that the crux of the incompatibility between Church

and theater lay in the radical differences between polytheistic and monotheistic attitudes

towards appearances, an epistemological issue with obvious implications for the

meaning of masking practices. Polytheistic traditions, Napier notes, exhibit a “tolerant

disposition toward the possible significance of appearances”: in “the cases where one

finds a genuine pantheon, there is, in the advent of unusual phenomena or events, a

natural tendency toward uncertainty and divination that at least in part results from not

knowing which member is to be held responsible” (MTP 14, 5). In a monotheistic

tradition, however, appearances are regarded as merely superficial in nature, ultimately

to be explained away:

A primary consequence of a world generated by a single omniscient force is that all change can be made accountable to that one ideation.... [A]ny unaccountable transition or inexplicable manifestation may be attributed in the end either to God or to Satan; traffic between supernatural forces and multiple conflicts among them become highly improbable.... For the mortal’s part, the understanding arose that beneath the guise of appearance, beneathprima facie evidence, was an innate reality that was not diverse, but devoid of unaccountable diversities. (MTP 5)

69 Charles Gallon, Personal Aspects of the Roman Theatre (Toronto, 1972), p. 37. Cited in MTP, p. 6.

104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The use of the mask in the classical theater was consistent with a polytheistic tolerance

for the ambiguity and mutability of the realm of appearances, indeed, John Jones notes

that for the Greeks, the dramatic mask was “known to have no inside”:

Its being is exhausted in its features. To think of the mask as an appendage to the human actor is to destroy the basis of the ancient masking convention by inviting the audience to peer behind the mask and demand of the actor that he shall cease merely to support the action, and shall begin instead to exploit the action in the service of inwardness.70

Rather than express the inward essence of the actor, masks were used to make manifest

transitions or transformations at the level of appearances: that is, to dramatically address

the question of mutability. In this, Napier points out, the use of the mask in the classical

theater is wholly consistent with the use of masks in other cultures and social contexts:

The presence of masks in situations relating to transition is so commonly the rule that exceptions to it are hard to find. Whether the change is from one social status to another or in the conscious states of the mask wearers or their audiences, again and again mask users or their observers or both attest to some change in conjunction with a mask’s presence. These transitions from one state to the next may occur in any number of easily recognizable categories, the most common being rites of passage, curative ceremonies and exorcisms, and religious and secular dramas. (MTP 16)

In light of the transformational significance of the mask across cultures, it is worth

noting that the Greek word for mask(pmsopon ) properly signified a manifestation or a

figure, and not simply a material thing; as Napier notes, “such a primary meaning

70 John Jones,On Aristotle and Greek Tragedy (London, 1971), pp. 44-45. Cited in MTP, p. 9.

105

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Implies that masks were conceived of as belonging to a much broader class of

phenomena than the mere object (mask)prosopon is usually taken to mean” (MTP 8).

Christianity’s Insistence on the mere superficiality of diverse appearances

militated against their acceptance of the classical notion personae.of Only for the devil

were apparent changes real. Anti-theatricality was therefore not an accidental, but an

essential, aspect of early Christian theological self-definition. This accounts for the

radical constriction and simplification of the meaning of the mask in Medieval drama:

Though the mask survived in several forms into the Middle Ages, interestingly what did not survive beyond the classical period was a sophisticated theater of personae, an extensive pantheon of mask types. Theatrical masking of the postclassical and early medieval eras was in fact far less organized and complex than in earlier times, so that when personae did appear, they did so in the standard plots common to the mystery plays and folk performances that may be witnessed in some European villages to the present day. In the absence of complex personae... masks naturally lost some of the easy familiarity given to them on stage and came more frequently to be associated with the sinister and the evil. (MTP 7)

Here we see the emergence of the limited and pejorative meaning that is so often

attributed to masks in modernity. Eventually, writes Napier, “the association between a

diabolical nature and a mask that disfigures and conceals came to exclude nearly all

other interpretations” (MTP 12).

At the same time, Christian writers were redefining the very meaning of the

Latin wordpersona itself. Boethius formulated the theological definition that would

eventually be adopted by the Scholastics: “naturae rationabilis individua substantia” -

an individual substance of a rational nature. The conceptual distance traveled between

106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mask and individual, appearance and substance measures the radical transformation

undergone by the classical notion of thepersona in the early centuries of the Church.

Napier suggests that we see this radical shift in the meaning ofpersona as part of a

general drive towards objectification that characterized the Church’s attitude towards a

number of different phenomena, including the Trinity and the Sacraments. In this light,

it is telling that the fundamentally dramatic explanation proposed by Sabellius to the

paradox of the Trinity would be rejected as heresy, for he argued that “the three forms

of divine presentation were mere forms and nothing more, that behind each mask there

stood individually the same actor, portraying in succession the roles of creation,

redemption, and sanctification.”71 Marcel Mauss also suggests that the historical

development of Christian beliefs about the unity of the human person were closely

linked to early debates about the unity of the persons of the Trinity, a theological issue

that was resolved by the Council of Nicaea (325 C.E.) only when (in response to the

perceived threat of Arianism) it decisively asserted the distinctness and

consubstantiality (homoousios) of the three divine persons. As Mauss puts it, unitas“ in

tres personas, una persona in duas naturas” (that is, unity of the three persons of the

Trinity, unity of the two natures of Christ).72 Modem personhood and Christian

monotheism, Mauss explains, are closely connected: “It is from the notion of the ‘one’

that the notion of the ‘person’(personne) was created - 1 believe that it will long remain

71 G.L. Prestige, God inPatristic Thought (London, 1936), p, i 13. Cited in MTP 7. 72 Marcel Mauss, “A category of the human mind: the notion of person;the notion of self,” trans. W.D. Halls, in The Category of the Person: Anthropology,Philosophy, History, eds. Michael Carrithers, Steven Collins, Steven Lukes (Cambridge: Cambridge Univ. Press, 1985), pp. 1-25. Subsequent references to Mauss’ essay will be given parenthetically with the abbreviation CP.

107

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. so - for the divine persons, but at the same time for the human person, substance and

mode, body and soul, consciousness and act” (CP 20).

The theological conception of the person as an enduring substance underlying

all apparent changes of state completed the Christian objectification of thepersona : it is

this transformation that enabled Augustine to attribute an entirely novel—and

historically momentous—importance to inferiority as the privileged space of mental and

spiritual life. For as we have already seen, he regarded the human mind as a private,

interior space fully formed even prior to the acquisition of language. According to

Augustine, this interior space was to be understood as the proper field of spiritual

cultivation. “Do not go outward,” as he put it inDe vera Religions, “return within

yourself. In the inward man dwells truth” (“Noli foras ire, in teipsum redi; in interiore

homine habitat veritas”).73 Considering the crucial role he accorded to inwardness for

the spiritual welfare of the person, it is no wonder that Augustine “viewed theater and

its personae as antithetical to true identity” (MTP 8; for examples of Augustine’s anti

theatricality, see Book 3 of the Confessions).

Augustine’s writings about the Christian person laid the ground for all

subsequent philosophical reflections on personhood, interiority and consciousness:

indeed, it is due to Augustine’s influence that these three terms can so easily appear to

refer to the same conceptual terrain in the first place. For the Augustinian turn inward

entailed a fundamental re-orientation of directionthe of spiritual and philosophical

73 Charles Taylor, Sources o f the Self: The Making o f the Modern Identity (Cambridge, MA: Harvard Univ. Press, 1989), p. 129. References to this text will be given parenthetically with the abbreviation SoS. Augustine cited on same page.

108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reflection: introspection and the first-person standpoint would from then on constitute

privileged ways to access truth, whether about oneself, the world, or God. Charles

Taylor, sees Augustine’s contribution as historically decisive:

It is hardly an exaggeration to say that it was Augustine who introduced the inwardness of radical reflexivity and bequeathed it to the Western tradition of thought. The step was a fateful one, because we have certainly made a big thing of the first-person standpoint. The modem epistemological tradition from Descartes, and all that has flowed from it in modem culture, has made this standpoint fundamental—to the point of aberration, one might think. It has gone as far as generating the view that there is a special domain of “inner” objects available only from this standpoint. (SoS 131)

Seen against the background of this Augustinian legacy, of which Wittgenstein was

clearly both conscious and critical, his later philosophy’s aspect-seeing model of

physiognomy shows us how “person” and “persona” might be conceptually re-united.

Like the mask of the classical stage, the face of Wittgenstein’s later philosophy has no

inside: his physiognomy takes us beyond the conceptual distinction between appearance

and essence (significantly, theInvestigations replaces the notion of essences with that of

family resemblances, a concept that clearly trades heavily on the special role that faces

play in human forms of life74). The mentality that we can quite simply see when we

74 Here is Wittgenstein’s introduction of the concept of family resemblances: “Consider for example the proceedings that we call ‘games’. I mean board-games, card-games, ball-games, Olympic games, and so on. What is common to them all?—Don’t say ‘Theremust be something common, or they would not be called “games’”—but look and see whether there is anything common so all.—For if you look at them you will not see something that is common toall, but similarities, relationships, and a whole series of them at that. To repeat: don’t think, but look!.... I can think of no better expression to characterize these similarities than ‘family resemblances’; for the various resemblances between members of a family: build, features, colour of eyes, gait, temperament, etc., etc., overlap and criss-cross in the same way.— And I shall say: ‘games’ form a family” (PI §66-67). The direct seeing he exhorts us to engage in (“don’t think, but look!”) is clearly the same kind developed throughout his yarious physiognomic reflections, and the common presence of faces is, of course, no accident. Dennett, in his essay “Real Patterns,”

109

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. adopt the right stance towards the faces of other people reflects a “dramatic idiom” that

we have quite naturally mastered. Read this way, we can discern the logic underlying

Wittgenstein’s choice of Augustine as a philosophical foil in theInvestigations, for we

might say that Wittgenstein’s version of physiognomy constitutes a modem philosophy

of the persona, overturning at once both Augustinian anti-theatricality and inferiority.

But I am not implying that we should (or could) alter the way we regard actual persons:

the point is to offer a truer account of our intuitive grasp of “folk psychology,” not to

change folk psychology as such (an impossible task, in any case). Rather, I believe that

philosophical “dramatists” of the mind, such as Quine and Wittgenstein, simply reveal

how persons always have been like personae to begin with, a truth the ancients may

have grasped but that we have since lost touch with. The modem personae of the

intentional stance, therefore, call for a fundamental rethinking of the concept of

theatricality: for, as we have seen, the historical development of the Christian soul was

predicated upon the denigration of the earlier personae of the classical stage, and the

effects of that corrosive anti-theatricality are still alive and well. Masks are still

commonly associated with mere superficiality and appearances; thetruth, on the other

hand, is to be found in depths and essences. As a result, our vocabulary to describe

surfaces can seem profoundly impoverished.

exhorts us to the same kind of direct seeing during a discussion of whether a particular (partially garbled) visual image (labeled “A”) can be said to contain a particular visualpattern that he calls “bar-code”: a real pattern in frame A isbar-code with 25% noise. And sometimes, we can simply tolerate or ignore the noise. From this perspective, a real pattern in frame A is simply:bar-code. But is bar-code really there in frame A? I am tempted to respond: Look! You can see it with your own eyes” (p. 103). Dennett’s point is that intentional!ty, like any pattern, notis a concrete fact, but nonethelessreal: at another point in the essay he calls folk psychology one of the “pattern-making perspectives we have on the buzzing blooming confusion that bombards us with data” (p. 104).

110

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. John Ashbery touches upon this very constellation of issues in his major poem,

“Self-Portrait in a Convex Mirror.” As Ashbery says to the face of Parmigianino in his

“Self-Portrait”: But your eyes proclaim / That everything is surface. The surface is

what’s there / And nothing can exist except what’s there” (190).75 Ashbery sees the

primary meaning of Parmigianino’sSelf-Portrait, especially its exquisite rendering of

the artist’s face, as having primarily to do with the question of personal identity, or

what—invoking theological categories—he calls the soul. “The soul has to stay where

it is,” Ashbery gathers from the depiction of the face in the painting: “It must move / As

little as possible. This is what the portrait says.” The soul must remain so still,

according to Ashbery, because it exists only in the mirroring (that is, speculative)

moment of one’s regard and attention. Put quite literally, Parmigianino could see his

face only while he looked straight into his reflection in the convex mirror. Any

movement of the face or eyes (or even attention) would have broken the specular

relationship between self and self. Understood metaphorically, Ashbery’s point is that

introspective reflection is at once master and slave of its own existence. So he reads the

painting as the record of a kind of incarceration that characterizes the existence (we

might say, is the condition of the possibility) of the soul in the first place: “the soul is a

captive, treated humanely, kept I In suspension, unable to advance much farther / Than

your look as it intercepts the picture.” Parmigianino’s painting shows that one’s

attention, and nothing else, is what lends being to the soul. Indeed, the painful “secret”

that the painting reveals is that, in fact, “the soul isnot a soul” at all (at least as

75 John Ashbery, Selected Poems (New York: Penguin, 1986).

I l l

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Christianity has conceived of it) since it has no substance or identity of its own: “The

secret is too plain. The pity of it smarts, / Makes hot tears spurt: the soul is not a soul, /

Has no secret, is small, and it fits / Its hollow perfectly: its room, our moment of

attention” (189). From a theological perspective, this knowledge is bitter indeed. But

seen differently, this “secret” (that the soul “has no secret,” since it is not some kind of

occult entity) is not painful at all, but rather liberating. This change in perspective is

registered by an alteration in poetic tone, the aqueous (and melodramatic) image of

spurting tears giving way to a jet of water playfully holding aloft the soul now pictured

as a bobbing ping-pong ball: “The whole is stable within / Instability, a globe like ours,

resting / On a pedestal of vacuum, a ping-pong ball / Secure on its jet of water” (190).

A ping-pong ball, of course, is hollow, nothing but surface all around. But Ashbery

does not rest content with this new, quite literally superficial, knowledge; perhaps

uneasy with the ease with which it was possible to assert that “everything is surface,” he

immediately puts into question our very ability to describe surfaces accurately: “there

are no words for the surface,” he concludes, “that is, / No words to say what it really is,

that it is not I Superficial but a visible core...” (190).

Ashbery apparently disdains a mere transvaluation of the psychological

distinction between inner and outer. That is, he cautions us against simply inverting our

traditional privileging of interiority in order to revel in a meaningless postmodern play

of surfaces. Our inherited conceptual schemes are not so easily overthrown, he

suggests. Indeed, he heaps scom upon those who would take this all too easy route,

calling them “beside the point”: “those assholes / Who would confuse everything with

112

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. their mirror games” ... “that would corrode the architecture / Of the whole in a haze of

suppressed mockery” (200). “They are out of the game,” he concludes, as if wiping his

hands clean, “Which doesn’t exist until they are out of it” (200). That life may be a

“game,” Ashbery implies, does not mean that it has no point, just as by denying the

interiority (or “secrecy”) of the soul he does not intend to deny its reality. The soul is

not an illusion, but a “visible core,” as he puts it. One can see why, after briefly toying

with a superficial conception of the self, he reconsiders and declares that “there are no

words for the surface.” For to speak of the soul as superficial would be to partake still

of a logic that opposes it to an essence. The two concepts go together and neither,

Ashbery suggests, captures the truth about the soul. As Nietzsche put a parallel point in

The Gay Science, tellingly invoking the figure of a mask: “What is ‘appearance’ for me

now? Certainly not the opposite of some essence: what could I say about any essence

except to name the attributes of its appearance! Certainly not a mask that one could

place on an unknown X or remove from it!”76 How, with the conceptual pictures we

have inherited, can we speak of the nature of the soul or the person without betraying

the truth one way or another? Ashbery’s feeling that “there are no words for the

surface” is understandable and apposite to our broader discussion of the historical fate

of the persona in the West: it is perhaps too easy for us to understand such families of

concepts as surface, theater and masks as opposed to depth, sincerity and faces. But I

hope this dissertation demonstrates that developments in the philosophy and science of

76 Friedrich Nietzsche,The Gay Science, Trans. Walter Kaufman (New York: Random House, 1974), p. 54.

113

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mind throughout the twentieth-century have been busy formulating the very words we

might use to characterize the “visible cores” that Ashbery rightly suggests we truly are.

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3. SeeingAs: Pictures, Perception andPersons

Like the sight of the human face itself, visual representations of the human

countenance have always exerted a special power upon those who behold them, whether

one prefers to characterize this power as religions or aesthetic (or some other way).

When it comes to understanding the presence or aura often associated with pictures of

the human face, one of the single most powerful and illuminating points was made by

E.H. Gombrich in his popular art history survey, The Story o f Art. At the beginning of

the first chapter (“Strange Beginnings”), devoted to the art of “prehistoric and primitive

peoples,” Gombrich takes a moment to discuss the magical beliefs “so-called primitive

peoples” have about images of all kinds, but especially those representing persons: “the

absurd feeling that what one does to [a] picture is done to the person it represents”(SA

40). Gombrich suspects that his readers may find such ideas radically alien, even

unintelligible. But he argues that, in fact, such “irrational” feelings are very much alive

and well within us, even if we may not be consciously aware of them. He insists that

we need to sympathetically understand such feelings if we hope to attain a proper

understanding of the function of images in prehistoric cultures: “these strange ideas are

important because they may help us to understand the oldest paintings which have come

down to us.”77 So he goes on to suggest a hypothetical exercise designed to elicit a

conscious first-person recognition of the “absurd” feelings he believes we still

77 E.H. Gombrich, The Story o f Art (London: Phaidon Press, 1995), p. 40.

115

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (unconsciously) have about images. Tellingly, the exercise he recommends focuses

particularly on a picture of a person’s face:

We cannot hope to understand these strange beginnings of art unless we try to enter into the mind of the primitive peoples and find out what kind of experience it is which makes them think of pictures, not as something nice to look at, but as something powerful to use. I do not think it is really so difficult to recapture this feeling. All that is needed is the will to be absolutely honest with ourselves and see whether we, too, do not retain something of the ‘primitive’ in us. Instead of beginning with the Ice Age, let us begin with ourselves. Suppose we take a picture of our favorite champion from today’s paper - would we enjoy taking a needle and poking out the eyes? Would we feel as indifferent about it as if we poked a hole anywhere else in the paper? I do not think so. However well I know with my waking thoughts that what I do to his picture makes no difference to my friend or hero, I still feel a vague reluctance to harm it. Somewhere there remains the absurd feeling that what one does to the picture is done to the person it represents. Now, if I am right there, if this queer and unreasonable idea really survives, even among us, into the age of atomic power, it is perhaps less surprising that such ideas existed almost everywhere among the so-called primitive peoples. In all parts of the world medicine men or witches have tried to work magic in some such way - they have made little images of an enemy and have then pierced the heart of the wretched doll, or burnt it, and hoped that their enemy would suffer. (SA 41)

Whatever one thinks of the surrounding discussion - and a certain amount of cultural

condescension admittedly colors Gombrich’s remarks - the moment where he asks us

how it would feel to poke out the eyes of a person’s photograph is pedagogically

brilliant: indeed, I recommend that my own reader take a moment to follow Gombrich’s

directions before reading on (I believe, however, that the impact of the exercise is more

powerful if one selects a photograph of a personal acquaintance rather than simply of a

public figure). Gombrich is, of course, right that most of us will not “feel as indifferent

116

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. about” poking out the eyes of the person in the photograph “as if we poked a hole

anywhere else in the paper.” This is a phenomenological fact of the utmost importance

for a clear understanding of the perceptual and cognitive dynamics that underlie the

making and beholding of visual representations of the human face. The strange and

unpleasant feeling (and the “vague reluctance”78) one naturally experiences while

poking out the person’s eyes in a photograph immediately makes clear what page after

page of abstract discussions about the face could never get across.

In this context, it’s worth considering the sense of “violence,” even “brutality,”

that is so frequently associated with the distorted physiognomies painted by Francis

Bacon. Keeping Gombrich’s point in mind, imagine, just for a moment, what the

visceral feeling associated with the disfiguration of the “mere” image of a face suggests

about how Bacon might have felt while painting a distorted portrait of an intimate

friend, such as his lover George Dyer or friend Isabel Rawsthome. Imagine, as well,

how the friends he portrayed in such a distorted way might have felt about it too. A

brief and fascinating exchange between Sylvester and Bacon speaks directly to this

issue. Explaining to Sylvester why he preferred painting portraits of people from

photographs and memory rather than in the presence of the person, Bacon said: “They

inhibit me. They inhibit me because, if I like them, I don’t want to practice before them

78 Gombrich’s expectations about our response to this exercise were empirically confirmed by a class of mine in which I assigned it. I asked the students in the class, which was on the subject of modernist portraiture, to each bring in a photocopy of a photograph of a person they cared about (friend, spouse, family member). In class, I then asked them to poke two holes in some insignificant area of the photograph. All the students quickly complied, surely wondering all the while what I was up to. I then asked them to poke out both eyes of the person in the photograph. Much consternation and hesitation ensued, with some students taking widely varying amounts of time to comply with my instructions, and with a number of other students ultimately refusing to finish the exercise (interestingly, many of those who refused were parents who had brought in a photograph of a child).

117

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the injury that I do to them in my work. I would rather practice the injury in private by

which I think I can record the fact of them more clearly” (IFB 41). In response to a

follow-up question from Sylvester (“In what sense do you conceive it as an injury?”)

Bacon reverses his position and implies that in fact the injury is real only from the

(mistaken) perspective of the person being portrayed: “Because people believe - simple

people at least - that the distortions of them are an injury to them - no matter how much

they feel for or how much they like you” (IFB 41). Luckily, Sylvester presses the issue,

and after asking Bacon whether he does not think this feeling on the part of his friends

is “probably right,” he receives a thoughtful and tentative acknowledgement from

Bacon: “Possibly, possibly. I absolutely understand this. But tell me, who today has

been able to record anything that comes across to us as a fact without causing deep

injury to the image?” (IFB 41).

A number of important insights into Bacon’s art can be gleaned from this

searching exchange. The multiple reversals in Bacon’s position during the course of the

conversation, however, need to be approached with great care. Rather than take his

reversals and self-contradictions as signs of inconsistency or equivocation, or worse yet,

choose one of his statements over the others as indicative of his true belief, I think we

need to take all of his remarks together at once - they are probing and uncertain

acknowledgements of many sides of a single, very complicated truth about the power of

figuration and disfiguration. One thing that the exchange makes clear is that both

Bacon and those he painted did indeed recognize an element of violence (that is,

“injury”) in his use of distortion. What he does when he distorts the images of people

118

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. is, in a way, injurious. And then again, of course, it is not: only a “simple” person

would think that. And yet again, “possibly” injury is in fact involved, though it is

necessary and done to theimage, not the person: after all, “who today has been able to

record anything that comes across to us as a fact without causing deep injury to the

image?” But consider, for a moment, how readily Bacon speaks here of “causing deep

injury” to an image, as if that made any more sense than the possibility of injuring

persons by distorting pictures of them (would one speak of “injuring” a stone, for

instance?). The perceptual and affective boundary between image and person is

profoundly ambiguous here. In fact, he ends up voicing a quasi-magical sentiment not

all that far from where he began in this particular exchange: “I don’t want to practice

before them the injury that I do to them in my work.”

Gombrich’s general psychological point about the contemporary existence of

“magical” image-related feelings is certainly confirmed: Bacon and his friends were

apparently quite familiar with (if also ambivalent about) the “queer and unreasonable”

idea that “what one does to the picture is done to the person it represents.” Indeed, I

suspect that the “feeling” to which Gombrich draws our attention has a profound

affinity to the “sensation and the feeling of life” that Bacon often refers to in so many

different ways as he tries to explain his art to Sylvester (IFB 43). As Bacon was fond of

saying, he wanted his images to directly strike the “nervous system” of those who

viewed them: explaining his use of distortion, he once said “it’s an attempt to bring the

figurative thing up onto the nervous system more violently and more poignantly” (IFB

12). I propose that whatever other significance Bacon’s references to violent “feelings”

119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. might have, one of them - and perhaps the most important - is the unpleasant “feeling”

that Gombrich calls “absurd,” which all of us have when we disfigure (or even

contemplate disfiguring) a representation of a person’s face. But this point, as

important as it may be as a beginning, only gets us so far, for we still need to understand

why we have this feeling when we disfigure images of persons. After all, no actual -

shall we say physical - injury is done to the person portrayed when Bacon distorts his or

her mouth or nose or eyes (“I don’t think it is damage,” as Bacon puts it later in the

same conversation). Yet is the “feeling” we are trying to put our finger on, as

Gombrich says it is, really “absurd”? How, in any case, can a feelingbe absurd? That

an idea (which is formulated as a proposition) can be absurd is understandable, but one

wonders whether this adjective can appropriately be used to modify the kinds of mental

states referred to by the word “feeling.” Indeed, it appears that Gombrich has been

eliding the distinction between ideas (which must be expressible in propositional form)

and feelings (which, at least, do notneed to take propositional form). The “queer and

unreasonable idea” Gombrich discusses (“that what one does to the picture is done to

the person it represents”) surely is absurd, but the vague, unpleasant feeling that

underlies this idea is not the same thing as the idea itself, and I would say is not absurd

at all. Indeed, one can have this feeling without it ever taking the propositional form of

the idea Gombrich refers to. Though the feeling can provoke the formulation of the

idea, the two are conceptually distinct, and each can exist in the absence of the other. It

is perfectly plausible that a person could hold the belief expressed in Gombrich’s

magical “idea” without that person ever experiencing the feeling beforehand, in the very

120

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. way that many people express articles of religious faith without ever experiencing any

corresponding first-person religious “feelings” of grace, absolution, enlightenment or

what have you. Just so, ideas about magic can be transmitted and absorbed simply by

means of linguistic communication rather than first-hand experience. While it is very

likely that there is an intimate - though historically and culturally contingent -

association between the feeling Gombrich speaks of and the magical idea he calls

“queer and unreasonable,” the two have nonecessary logical connection to each other.

We do not need history in order to explain the contemporary presence of this feeling; in

fact, I would go further and say that it is our ability to experience this feeling in the

present that enables us, in the first place, to grasp its cultural function in the past.

It is because Gombrich fails adequately to distinguish between idea and feeling

that he ever mistakenly labels this feeling “primitive” in the first place and then

proceeds to wonder at its “survival” into the “age of atomic power.” If theidea

Gombrich discusses may very well be primitive, the feeling is most definitely not. If

we found modem people who expressed absolute certainty in the belief that one can

harm others simply by disfiguring pictures of them, that would be something to marvel

at. On the other hand, I would argue that the odd feeling we all experience when we

consider poking out the eyes of a person’s photograph (publicly observable in the

hesitation we manifest while contemplating the unpleasant task) is simply part of our

natural biological and cognitive makeup: I suspect Bacon was quite close to the literal

truth when he insisted on the importance of our “nervous systems” to the effect of his

art. But it’s still not clear why we have this feeling. After all, as we have already

121

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. noted, we are not gouging out the eyes of a living person who can experience pain and

suffering: physical revulsion at the thought of that bloody act would require little

explanation. And clearly we do not perceptuallymistake a picture for a person, for then

the quality and intensity of our feeling would be the same in both instances. So the

reason why we might experience a feeling of “vague reluctance” when contemplating

“injury” to the eyes of a person in a pictorial representation is not self-evident.

Wittgenstein’s discussion of aspect-seeing can help us here. Consider, for example, his

comments about our ability to see mentality in the eyes of other people, which we

discussed in chapter two:

As we noted when we first cited this passage, Wittgenstein is simply acknowledging the

everyday fact that we do not see eyes - as we would if we conformed to empiricist

theories of sense perception - as purely passive organs of sensory reception. Rather, we

clearly recognize in them the active life of the mind: eyes, that is to say, flash, radiate,

and gleam (and Wittgenstein’s point is that these are not mere figures of speech but

rather perceptual descriptions). But this is not because such aspects of mind are really

there, just like any other factual trait of the eyes. As Wittgenstein insists, the “glance”

is not an empirical fact like the color of an iris, or the dilation of a pupil: “I should

contradict anyone who told me I saw the glance ‘just the way’ I see the shape and

122

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. colour of the eye” (Z 223). If we can see mentality in the appearance of a person’s

eyes, that is simply because of the recognitional attitude we naturally take towards

them. The “look in the eye,” in other words, is an aspect of their appearance, not a

physiological fact.

But because there is no fact of the matter in the physical eyes themselves that

determines our recognition of mentality in their appearance, it follows then that there is

no reason why we should not be able to recognize such aspects of mind in pictorial

representations of eyes as well. All of this, of course, applies equally to the face as a

whole. Recognition of mentality in thepicture of a face, however, is additionally

colored by a simultaneous recognition that we are beholding a representation, and not

an actual face. But as Gombrich’s point about newspaper photographs and Bacon’s

remarks about “injury” make clear, the awareness of the difference between seeing an

actual person and a mere representation of a person can, at times, become experientially

blurred or problematic. Ideas are one thing, our feelings another. In other words,

though we consciously know that we are seeing a personin a picture, we may at the

same time experience a vague sense of seeing the pictureas a person too. Consider for

instance, Velazquez’s portrait of Innocent X, a masterful painting that somehow

provokes this kind of perceptual blurring (unfortunately, I suspect that if one does not

immediately see the truth of this claim when looking at Velazquez’s painting, then no

amount of argument will help). By stressing the phenomenological difference (and

occasional confusion) between seeing “in” and seeing “as,” I am of course invoking -

but also marking a disagreement with - Richard Wollheim’s well-known philosophical

123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. account of the distinction between these two Muds of visual perception in the second

edition of hisArt and its Objects (“seeing-in” then goes on to serve as the theory of

representational seeing in his magisterialPainting as an Art).79 In the first edition of

Art and its Objects, Wollheim had employed Wittgenstein’s notion of seeing-as from

section xi of theInvestigations as a general theory for the visual perception of pictorial

representations. In the second, and expanded, editionArt and of its Objects, however,

Wollheim came to view his earlier use of seeing-as as incorrect, and argued instead for

another closely related phenomenon that he called “seeing-in” as a better way to

characterize the type of perception we appropriately bring to pictorial representations.

As he explains, in a chapter entirely devoted to marking the conceptual distinction

between seeing-in and seeing-as: “Where previously I would have said that

representational seeing is a matter of seeing x (= the medium or representation)y (=as

the object, or what is represented), I would now say that it is, for the same values of the

variables, a matter of seeing y in x” (AO 209). According to Wollheim, in other words,

it is a conceptual mistake to claim, as I - along with Gombrich and Bacon - ventured a

moment ago, that one ever sees a pictureas a person.

Wollheim’s discussion of the differences between these two forms of seeing

raises a number of distinct issues, yet I think what is at stake in the discussion as a

whole is neatly captured by what he calls the “twofold thesis”: “The thesis says that my

visual attention must be distributed between two things though of course it need not be

equally distributed between them” (AO 213). The two “things” Wollheim is referring

79 See Richard Wollheim,Art and its Objects, ch. 5.

124

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to here are, on the one hand, the “medium or representation” and, on the other, “the

object, or what is represented.” According to Wollheim, seeing-in respects the

phenomenological difference between these two components of representational seeing,

whereas seeing-as does not, the latter, he claims, dissolving an awareness of the

medium Into an all-consuming perception of the thingrepresented. In other words,

according to Wollheim, when I see x y,as I am only conscious of y, and no longer of x,

though I can consciously bring my attention back to x if I wish (then losing sight of y).

But Wollheim points out that when I see a face in an oil painting, I amsimultaneously

aware of the face represented and the various markings that make up the medium of the

representation (though, depending on one’s state of mind and the kind of handling

exhibited by the medium, the distribution of attention between these two aspects of a

representation may vary: looking at Ingres’ portrait of Louis-Francois Berlin I may

attend more to the face represented, while seeing Matisse’s portrait of his wife with a

green stripe, I may very well attend more to the medium of representation).

The major target of Wollheim’s argument is Gombrich’s account of perception

inArt and Illusion (“I have argued for [the twofold thesis] in arguing against

Gombrich,” as he acknowledges):

For it is a central thesis ofArt and Illusion that, in looking at representational pictures, I am incapable of this kind of twofold perception. Gombrich attempts to clinch the point by assimilating what he calls the ‘seeing canvas’/’seeing nature’ disjunction, or what I have expressed as seeing the medium versus seeing the object, which holds for picture perception generally, to the seeing the duck/seeing the rabbit disjunction, which holds for the special case of looking at the ambiguous duck-rabbit picture. Everyone would recognize that the second

125

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disjunction is exclusive, or that we cannot simultaneously see the duck and the rabbit in the picture, and through assimilating the two disjunctions Gombrich is able to claim that the first disjunction is exclusive too. I cannot be simultaneously aware of the medium and of the object of the representation and to perceive both I have to switch perception. (AO 213-4)

Of course, Wollheim is right to oppose any theory of picture perception that would deny

the possibility of a simultaneous awareness of medium and object, to use his own

terminology. And he is surely correct to dispute Gombrich’s assimilation of the duck-

rabbit disjunction to visual perception in general: only particular types of perceptual

objects exhibit the ambiguity shown by Jastrow’s duck-rabbit. Indeed, Wollheim need

only have noted that even the duck-rabbit picture exhibits twofoldness: while one

cannot simultaneously see the duckand the rabbit, one can see the duck aspect while

remaining aware that one is seeing that aspectin a drawing, and likewise with the rabbit

aspect. But my primary concern is whether in having achieved such conceptual clarity

- firmly marking off seeing-in from seeing-as so as to preserve the twofoldness of

pictorial representations - Wollheim does not blind himself to the kind of

phenomenological datum about our relationship to images that Gombrich’s exercise in

The Story o f Art so vividly brought to our attention. How would Wollheim account for

the “feeling” Gombrich and Bacon each refer to, except by denying its possibility

altogether?

On this score, it is interesting to see that Wollheim chooses to discuss a portrait

in order to demonstrate what is wrong with Gombrich’s account of perception. Arguing

that the twofold thesis, unlike Gombrich’s view, can respect “what is distinctive

126

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. phenomenologically... about seeing something or someone in a representation,”

Wollheim offers the following illustration, which implicitly rales out the possibility that

we might see a picture as a person:

[The twofold thesis] tells us what is experientially different, for example, seeing Henry VIII in Holbein’s portrait, as opposed to seeing him face to face. The suggestion favoured by Gombrich, which is that, while looking at Holbein’s portrait, I can always stop seeing Henry VIII, switch my perception, and then be visually aware of the canvas, clearly does not fill this need. For instead of citing some actual characteristic of a present experience, it merely invokes the possibility of an alternative experience, and this is not phenomenology. Indeed, it is Gombrich’s failure to assign to the seeing appropriate to representations a distinctive phenomenology that impels him towards the view that there is nothing distinctive about the seeing of representations, or that seeing someone’s representation is quite continuous with seeing that person face to face - with all that such a view implies or suggests. (AO 214-5)

Now, it is clear that Wollheim is again right when he notes that any failure to

distinguish between actual face-to-face perception and the perception of a face in a

painting is conceptually faulty. As I stated earlier, recognition of mentality in the

picture of a face is necessarily colored by the awareness that the recognition is in

response to apicture. Otherwise, one would feel no differently whether cutting into an

actual person or a photograph of that person. In other words, art is not illusion, as

Wollheim rightly insists (but as does Gombrich as well). However, Wollheim’s account

is unable to account for the phenomenological “fact” that poking out the eyes in a

photograph of a person will not feel the same as poking two holes in the middle of a

blank sheet of paper. Much of Wollheim’s discussion is absolutely convincing and

helps advance our understanding of representational seeing, especially since it explicitly

127

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. accounts for the awareness of artifice that is involved when we look at pictures, but

seeing-in and seeing-as are not as phenomenologicaily distinct as Wollheim’s

conclusions suggest. In this respect, I would say that while Gombrich’s explanations

may not be as philosophically acute as those of Wollheim, the former nonetheless show

an intuitive feel for an important aspect of our experience of images that Wollheim’s

arguments seem inadequately sensitive to (perhaps the difference in their sensibilities

accounts for the fact that it is Gombrich who has produced some of the most important

studies yet of physiognomic expressions in the literature on the psychology of art80).

Gombrich, that is, rightfully acknowledges that the distinction between persons

and pictures can, at times, become phenomenologicaily confused, however difficult and

problematic this acknowledgement makes our understanding of pictorial representation

in general. The problem such perceptual confusions introduce into our theories of art, I

would suggest, are not simply the result of misformulated theories (on the part of

Gombrich) but rather evidence of something essential to the nature of pictorial

representation and perception. That Gombrich’s explanation is partly mistaken does not

show that the problem he draws our attention to is not real: on the contrary, he has put

his finger on a fact that must be adequately accounted for in any comprehensive

psychology of art. In fact, I would call it a serious fault of Wollheim’s otherwise

powerful theory of art that the best “solution” he can suggest is to deny the reality of the

problem in the first place. Therefore, I would reproduce one of Wollheim’s criticism of

80 See Art and Illusion, ch. 10 (“The Experiment of Caricature”) and “The Mask and the Face: the Perception of Physiognomic Likeness in Life and in Art”Art, in Perception, and Reality, Ed. Maurice Mandelbaum (Baltimore: Johns Hopkins Univ. Press, 1970), pp. 1-46.

128

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gombrich’s theory, though this time taking it as a summation of one of Gombrich’s

most valuable insights: that “seeing someone’s representation is quite continuous with

seeing that person face to face - with all that such a view implies or suggests.”

Continuous, however, not identical. It is only because there is such a continuity

(“absurd” as it may sound) that a pictorial representation of another human being can

face us with a quasi-moral injunction not to do harm (evident in our “vague reluctance”

to poke out his or her eyes). This does not mean that such continuity between image

and person is incompatible with the “twofold” awareness upon which Wollheim insists:

on his sense of the importance of “twofoldness” to a proper understanding of

representational seeing, I completely agree. As Wollheim argues, Gombrich is surely

mistaken to categorically deny its possibility. But I would suggest that there is

something wrong with Wollheim’s account as well, for it appears that the

phenomenological qualities he associates with seeing-in and seeing-as can actually co

exist in a single experience of representational seeing - at least, we may need to add,

when the representation we are seeing is of a human face. Faces, it appears, are in a

representational category all their own. It follows, then, that we may need a theory of

visual perception specific to seeing representations of faces. There is, after all, no

reason to think that an absolutely general and homogeneous theory of representational

seeing is even possible, let alone desirable. It may be the case that certain types of

representational objects call upon different kinds of seeing than do others (real faces

certainly do, as cognitive neuroscience has shown us). We may naturally have different

kinds of perceptual relationships to different kinds of represented objects. Wittgenstein

129

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. seems to have suggested as much when he said, “You need to think of the role which

pictures such as paintings (as opposed to working drawings) have in our lives. This role

is by no means a uniform one” (PI 175). Pictures of faces, at least, appear to elicit a

.kind of perception that is special, and it seems also especially difficult to understand.

Indeed, perhaps Wollheim’s mistake was simply to attempt to specify a single form of

seeing that would encompass all instances of representational seeing in general.

But I believe another problem with Wollheim’s account is linked to the way he

reads section xi of thePhilosophical Investigations, the text from which he derived his

original account of seeing-as. Indeed, one senses that something may be wrong with

Wollheim’s interpretation of section xi as soon as he claims that “seeing-as” is

fundamentally related to what he calls the “straightforward perception” of objects

present to the senses. Because so much depends on the cogency of Wollheim’s reading

of Wittgenstein here, I will quote his interpretation at length;

... seeing-as draws upon no special perceptual capacity over and above straightforward perception. Rather it partially is, partially is a development out of, an aspect of straightforward perception. The aspect is this: Whenever I straightforwardly perceive something, which ex hypothesi is present to the senses, my perception of it is mediated by a concept, or in perceiving it I subsume it under a concept. For anyx, whenever I perceive x, there is always some/such that I perceive x as/. But it is crucial to an understanding of seeing-as to recognize that my seeingx as /is not just the conjunction of my seeingx and my judging it to be/. Such a view, which has gained currency among perceptual psychologists who talk of perception as hypothesis, errs in that it leaves the judgment external to the perception. It was just this view that Wittgenstein tried to combat when he asked us to consider cases where we switch from seeing something or other as this to seeing it as that. For the relevance of such cases is that they allow us to observe how experience and concept change not

130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. merely simultaneously but as one. It is a misfortune of Wittgenstein’s exposition of his argument that he chose as examples of alternating perception cases of alternating perception of representations: notably, the duck-rabbit drawing. For such cases introduce additional complexities, which can be the source of confusion. But the fundamental point in Wittgenstein’s argument, which remains, is that, when I seex as/,/permeates or mixes into the perception: the concept does not stand outside the perception, expressing an opinion or conjecture on my part about x, and which the perception may be said to support to this or that degree. (AO 219-220)

I will begin by saying - in order to immediately establish the nature of my interpretive

disagreement with Wollheim - that “the fundamental point” or, better yet, motivating

concern of Wittgenstein’s discussion in section xi is to understand how we can see

mentality in the appearance of a face (as well as other phenomena) and not, as

Wollheim claims, simply to argue that “when I see x as/,/permeates or mixes into the

perception.” Certainly, the point that Wollheim emphasizes is a crucial part of the

larger project of section xi (as confirmed by our discussion in chapter two), but itonly is

a part: that is, Wollheim does not misread Wittgenstein (or at least the part he reads),

but he ultimately fails to see the underlying motivation orpoint of the whole discussion.

Wittgenstein’s “fundamental point” is not to explain how we see what is there, but to

understand how we see what cannot be there, as a fact of the matter, that is (such as the

mental states of timidity, sadness, or courage): after all, as philosophers of mind

working in the field of developmental psychology have pointed out, “there is nothing

specifically menial about human faces” (CPM 253).

I should note, however, that Wollheim’s “straightforward” way of approaching

section xi is not uncommon. In fact, few of Wittgenstein’s readers (even very careful

131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ones) have grasped the motivating importance of physiognomy and the face to the

discussion of aspect-seeing in theInvestigations, perhaps because this understanding

requires a grasp of the genesis of Wittgenstein’s concern with faces in his reading of

Weininger’s physiognomic treatise,Sex and Character, and his subsequent study of

Spengler’s “physiognomic flair” inThe Decline o f the West.®1 Close reading of the

Investigations alone, in other words, may not be enough. As noted in chapter two, it is

from Weininger’s physiognomic remarks about the portraits of men of genius that he

borrowed the term “aspect” in the first place, and it is by reading Spengler’sDecline

that he realized the methodological superiority of morphological description over

systematic theorization. Indeed, an incident recollected by Wittgenstein’s friend Miles

Drury shows how much a part of his everyday thoughts the ideas of Spengler eventually

became. One day in the autumn of 1930, Drury reports, Wittgenstein came to visit him

after strolling through Cambridge and declared, with great distress, that he had just seen

a pictorial representation - tellingly in the form of a series of portraits - of Spengler’s

theory of cultural decline:

I was walking about in Cambridge and passed a bookshop, and in the window were portraits of Russell, Freud and Einstein. A little further on, in a music shop, I saw portraits of Beethoven, Schubert and Chopin. Comparing these portraits I felt intensely the terrible degeneration that had come over the human spirit in the course of only a hundred years.82

811 again refer the reader to Ray Monk’s magisterial biography of Wittgenstein,The Duty o f Genius, for a meticulous account of the importance of both Weinioger and Spengler to Wittgenstein’s whole approach to philosophy, and indeed life. For instance, Monk notes that in 1931, “references to Weininger and Weiningerian reflections abound in Wittgenstein’s notebooks and conversations. He recommendedSex and Character to his undergraduate friends, Lee and Drury, and to [G.E.] Moore” (312). He likewise recommendedDecline o f the West to his friends, even after Spengler’s association with the National Socialists rendered it a politically suspect work to many (cf. p. 315). 82 Rush Rfaees, ed., Recollections of Wittgenstein (Oxford: OUP 1984), p. 112. Cited in WDG 299.

132

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Portraiture is considered the most important and culturally characteristic kind of

Western (or Faustian) art inThe Decline o f the West: “Faustian man... found the most

genuine, the only exhaustive, expression of his life-feeling inPortrait'' the (DoW 136,

italics in original). And, Wittgenstein could not have thought of Spengler and

portraiture without recalling Weininger’s remarks on portraits and genius. InSex and

Character, Weininger observes that “men of genius” will exhibit “incredibly great

changes” in “personal appearance... from time to time,” a physiognomic indication of

great mental power that can be verified simply by looking at their portraits:

“Comparisons of the portraits at different times of Goethe, Beethoven, Kant, or

Schopenhauer are enough to establish this” (SaC 65). It is this that leads Weininger to

conclude that the “number of different aspects that the face of man has assumed may be

taken almost as a physiognomical measure of his talent” (SaC 65). It is likely this very

passage that Wittgenstein has in mind when, in early drafts of §§536-537 of the

Investigations where he first considers the timid or courageous face, he explicitly cites

Weininger in connection with the idea of an “aspect of the face itself’ (cf. the draft

version published in thePhilosophical Remarks, for example, which preserves the

parenthetical citation of Weininger). Aspects are, therefore - and from the moment of

their genesis in Wittgenstein’s philosophical reflections - a physiognomic concept.

And seeing-as is, first and foremost, a seeingaspects: of seeing a face as fearful, for

example. One can now see how misplaced is Wollheim’s complaint that it is a

“misfortune of Wittgenstein’s exposition of his argument that he chose as examples of

133

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. alternating perception cases of alternating perception representations:of notably, the

duck-rabbit drawing.” Far from being a “misfortune,” Wittgenstein’s choice of such

visual objects as duck-rabbit drawings for his exposition reflects the pictorial origins of

his interest in aspects in the first place: aspects of mind exhibited in portraits. The

representational nature of such examples is not somehow extrinsic to Wittgenstein’s

concern with aspects - a factor that would merely “introduceadditional complexities,”

as Wollheim puts it - but rather present from the very beginning in Wittgenstein’s

reading of the discussion of portraits and physiognomySex in and Character.

Of course, Wittgenstein radically alters Weininger’s notion of physiognomy,

reversing Weininger’s insistence on a parallelism between inner and outer in favor of

the radical visibility of mental states. But even if he did not follow him on every point

(and in fact, reversed him on most) Wittgenstein knew clearly, and candidly

acknowledged, how much he owed to Weininger’s inspiration. When he received a

cool response after warmly recommendingSex and Character in 1931 to his colleague

G.E. Moore, Wittgenstein wrote to Moore in terms that show not only the intensity of

his feeling for Weininger but also the complicated nature of his relationship to his ideas:

It is true that he is fantastic but he is great and fantastic. It isn’t necessary or rather not possible to agree with him but the greatness lies in that with which we disagree. It is his enormous mistake which is great. I.e. roughly speaking if you add a ~ to the whole book it says an important truth, (cited in WDG 313)

It was by “negating” the inner-outer picture that underlay Weininger’s physiognomy

that Wittgenstein was able to grasp the “important truth” his physiognomic ideas

expressed. Many elements in Wittgenstein’s later writings, such as Part II of the

134

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Investigations, the Remarks on the Philosophy of Psychology, and much ofZettel, can

profitably be read as late flowerings of the conceptual seeds first planted in

Wittgenstein’s mind by his study of Weininger, and then Spengler. But only by

adopting a genetic and developmental approach to Wittgenstein’s later philosophy can

we see how important physiognomy actually was for him.

Another reason that it is difficult to see the importance of the face to the later

writings is that physiognomy was not primarily anobject of investigation, but rather the

source of Wittgenstein’s methodologicalorientation. He suggests as much in citing

Spengler during his essay on Frazer’sGolden Bough:

For us the conception of a perspicuous presentation is fundamental. It indicates the form in which we write of things, the way in which we see things. (A kind of Weltanschauung that seems to be typical of our time. Spengler.) This perspicuous presentation makes possible that understanding which consists just in the fact that we “see the connections.” (WDG 311)

In other words, physiognomy (that is, Spengler’s “physiognomic flair”) shapedhow he

came to think, but was far from the only thing he was interested in thinking about:

understanding how we recognize the “familiar physiognomy[Gesicht] of a word,” as he

tellingly put it, was certainly as important as understanding how we recognize the

expression of a timid face (PI 186). Yet, as we stated earlier, Wittgenstein could

express his understanding of how we experience the meaning of a word by likening a

word to a face (Gesicht) only because he already regarded faces physiognomically.

Without a sense of the continuity of Wittgenstein’s interest in physiognomy, the

scattered mentions of faces in section xi can easily seem but one kind of example

135

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. among many (no more important than duck/rabbit drawings, for instance). But the

other phenomena discussed in section xi (words, duck/rabbits, triangles, melodies,

portraits, etc.) are linked to, and emerge out of, a more fundamental interest on

Wittgenstein’s part in the recognition of mental states in the perception of faces (i.e.

physiognomy). Physiognomy can explain the meaning of these nonface phenomena

because, as he put it earlier in the Investigations , meaningis a physiognomy (“Die

Bedeutung eine Physiognomic,” PI 568). As I suggested earlier, however, I think it

would be quite difficult to see this simply by closely reading section xi on its own:

perhaps one must situate the discussion of seeing-as and aspect-seeing within the wider

context of Wittgenstein’s intellectual development in order to “see the connections” I

am gesturing towards.83 This, I suspect, is why Wollheim does not clearly grasp the

“fundamental point” of section xi himself.

In any case, only a failure to see the ultimate importance of mentality in general

to the whole of section xi could lead Wollheim, a little later in his discussion, to choose

a perception of an oak tree in order to illustrate how seeing-as works: “The concept/

enters the mind along with the perception x,of blends with this perception, and stays in

the mind to form the belief that x is/. So I look out of the window of a train and see a

tree which I straightaway see as an oak, which I thereupon believe it to be” (AO 220).

Wollheim’s point is not incorrect, but by using a “straightforward” and trivial object

83 Indeed, as I noted in chapter two, it is John Koethe’s study,The Continuity o f Wittgenstein’s Thought, that led me to search out these connections in the first place. In that text, Koethe even tentatively noted “a faint parallel with seeing Jastrow’s figure as either a duck or a rabbit in the discussion in Investigations, 537 of the possibility of seeing a face as timid or as courageous” (CWT 113). I hope it is now clear why Koethe is absolutely correct to see these moments as connected.

136

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. such as an oak tree in order to illustrate the nature of seeing-as, he minimizes the

conceptual novelty and importance of section xi, the point of which is to understand

precisely those instances of perception that are not at all straightforward, such as the

fearful expression of a face, the sadness of a musical melody, the meaning of a word, or

the striking likeness between two different faces. If Wollheim’s account of seeing-as

was all it really amounted to, one would not wonder at his decision to abandon it as

hopelessly inadequate to the understanding of art.

He comes closer to suggesting the conceptual reach of Wittgenstein’s theory of

seeing-as, however, when he discusses instances of seeing in which we attempt to see

some x as an /w e think or know the jc actually is not: for example, a line of trees as a

row of pirates; a church as an overturned footstool; or a mountain range as a naked

woman’s body (all Wollheim’s examples; AO 222). Here, however, Wollheim actually

misinterprets Wittgenstein, rather than simply failing to see his underlying point.

According to Wollheim’s explanation of the mental processes related to such “pretend”

forms of seeing-as, the mind actively imagines what would have to change aboutx in

order for it to be seen or experienced as/: “we must be able to imagine howx would

have, or would have had, to change or adapt itself in order to take on the property of

being/. We would have to be able to imagine just how much ofx would have to go,

and just how much could stay, under this transformation” (AO 222). Anyone who has

witnessed children playing pretend will know that this is a mischaracterization of the

kind of seeing that occurs under conditions of pretence. To see a bananaas a telephone

does not require that the child engage in any kind of imaginary process of visual

137

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. alteration or adaptation: would a child really have to grasp how much of a banana

would have to go, how much stay, before the child could see a bananaas a telephone?

Putting it this way immediately shows how overly intellectualized Wollheim’s account

is. The transformation of a banana into a telephone is notvisual but conceptual, ft is

appropriate for us to say that the child sees the banana as a telephone, but not because

he imagines the banana as lookinglike a telephone. Indeed, developmental

psychologists, as we noted earlier, have found that pretence operates by decoupling

(rather than altering or adapting) primary and secondary representations: as Karmiloff-

Smith and Russell explain, “pretending that a simple block of wood has a steering

wheel, a horn and four wheels in no way detracts from toddlers’ understanding of the

real properties of the block of wood and of real cars, nor does it change their

representations of such properties” (CPM 254). According to Wollheim’s account, on

the other hand, during the time a toddler pretended to see a block of wood as a car, he

would not be able to see the wood underlying the pretence: he would only be able to see

it as a car. Seeing-as, after all, is supposedly incompatible with twofoldness. And in

the following discussion, Wollheim follows his methodical reasoning all the way to the

unpersuasive conclusion that seeing-as necessarily precludes twofoldness:

We have already noted that, in those cases [such as the oak tree] where x is indeed believed to be/, seeingx as/goes beyond merely seeing x and simultaneously judgingx to be/: seeing x as / is a particular visual experience of x. So in the same way, in the case where x is not believed to be /, or is even believed not to be /, seeing x as/[such as a line of trees as a row of pirates] goes beyond seeingx and simultaneously imagining it to be/: it also is a particular, though a very different, experience of x. Now, just because it is, there is no room when seeing x as counterfactual/

138

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. also to be visually aware of those properties of x which would have to change if x were actually to be or become/. In other words, the properties that sustain my perception of x as/w ould have themselves to be perceptually masked if I am - as I have put it - to try out the appearance of/upon x. So twofoldness in the case of seeing-as is ruled out. (AO 222-223)

Key phrases such as “were actually to be or become” and “to try out theappearance o f f

upon x” betray what is wrong with Wollheim’s interpretation. He seems to believe that

seeing x as “counterfactual / ’ (the way he characterizes the/w e know x actually is not)

would entail some form of alteration in an inner, or phenomenal, picture of x (which

then perceptually “masks” the “properties that sustain my perception of x a s /’). But,

oddly enough, one of the most important elements of section xi is Wittgenstein’s

argument against the inner picture theory of visual perception: ‘The concept of the

‘inner picture’ is misleading, for this concept uses the'outer picture’ as a model; and

yet the uses of the words for these concepts are no more like one another than the uses

of ‘numeral’ and ‘number’. (And if one chose to call numbers ‘ideal numerals’, one

might produce a similar confusion.)” (PI p. 167). Wittgenstein especially cautions

against placing the “organization” of the visual impression on a level with color and

shape, which Wollheim surprisingly appears to do when he states that we must “be

visually aware of those properties of x which would have to change if x were actually to

be or become/.” Wittgenstein’s response to this kind of theory, which we have already

cited in chapter two, is clear about the conceptual problems it entails: “If you put the

‘organization’ of a visual impression on a level with colours and shapes, you are

proceeding from the idea of the visual impression as an inner object. Of course this

139

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. makes this object into a chimera; a queer!y shifting construction. For the similarity to a

picture is now impaired” (PI p. 168). Wollheim’s “x seen as counterfactual/’ can

certainly be called a “queerly shifting chimera,” as it entails the transformation of the

appearance of x along with a perceptual masking of those properties ofx which sustain

the transformation into the appearanceoff. Indeed, the implausibiiity of Wollheim’s

conclusion provides strong confirmation of Wittgenstein’s argument that the inner

picture theory leads inevitably to mistaken, even incoherent, conceptual conclusions.

I suspect that Wollheim’s discussion of pretence seeing may have been

prompted by the following discussion of Wittgenstein’s in section xi about children’s

pretend play, which, if not carefully read, could easily be taken to support Wollheim’s

incorrect point about “perceptual masking”:

Here is a game played by children: they say that a chest, for example, is a house; and thereupon it is interpreted as a house in every detail. A piece of fancy is worked into it. And does the child nowsee the chest as a house? “He quite forgets that it is a chest; for him it actually is a house.” (There are definite tokens of this.) Then would it not also be correct to say he sees it as a house? (PI 176)

The answer to both questions is clearlyyes: the childsees the chest as a house.

However, Wittgenstein’s point here must be carefully specified and delimited.

Wollheim’s argument above about x seen as counterfactual / would find support in this

passage if one read the interlocutor’s declaration that the child “quite forgets that it is a

chest” as evidence that Wittgenstein too believes the child is no longer aware of the

properties of the chest that sustain his “perception ofx [=chest] as/[=house].” It is,

however, always dangerous to attribute remarks of the interlocutor unambiguously to

140

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Wittgenstein himself. For here, Wittgenstein is only agreeing that one could and should

say of the child that “for him it actually is a house”: but there is absolutely no reason to

go further and take Wittgenstein’s use of “see” here as a claim about the actual visual

perception experienced by the child. Indeed, one cannot imagine Wittgenstein

venturing that type of claim in the first place (recall the “Inner picture” argument): what

matters is not whether the child quasi-hallucinates the chest into a house, but that, for all

intents and purposes, one can see the child’s behavior as criterial of the fact that he is

regarding the chest as a house, period. A “piece of fancy” (that the chest is a house) has

been worked into the chest, but there is no reason to think the imagined “house” has

eclipsed or masked the appearance of the “chest” altogether. The transformation is

conceptual, not visual, though the only way to accurately capture this transformation in

language is to say that the childsees the chest as a house. After all, that’s how he treats

it: “there are definite tokens of this.” (I would encourage my readers to take a moment

and try pretend seeing-as on their own, in order to see for themselves which account

better describes its phenomenology.)

That Wollheim’s interpretation of seeing-as is inaccurate, however, does not in

any way vitiate his account of representational seeing in a text suchPainting as as an

Art, for Wollheim assimilates to seeing-in much of what properly belongs to seeing-as

in the first place. This conceptual assimilation explains why the phenomenological

boundary he attempted to establish between the two seemed problematic and at times

confused, for the attributes he associates with each are in actuality but aspects of one

(very internally diverse) form of seeing: that is, seeing-as or, better yet, aspect-seeing

141

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (but the quasi-technical terms themselves are not important). It appears Wollheim

mistakenly felt compelled to discard the theory of seeing-as in section xi of the

Investigations simply because he held an overly narrow, and therefore skewed,

conception of what seeing-as was like in the first place.

In other words, twofoldness notis necessarily ruled out by seeing-as. Seeing a

picture of a face as expressive of mind (as one can easily do when seeing the face in

Velazquez’s Innocent X, for example) is perfectly compatible with a simultaneous

awareness of the representational medium that supports our ability to recognize the

Pope in the image in the first place. A consciousness of pretence and artifice, that is,

naturally and inevitably accompanies our experience of the recognition of mentality in

representations of faces. It might be better, however, to say that pretence is what

enables us to see a picture as a person to begin with. Touching on this issue in section

xi, Wittgenstein introduces what he calls a “picture-face”:

Fig. 4

He then explains that, “In some respects I stand towards it as I do towards a human

face. I can study its expression, can react to it as to the expression of the human face.

A child can talk to picture-men or picture-animals, can treat them as it treats dolls” (PI

166). Wittgenstein’s point is clearly that pretence, just like the imaginary play of

children, is what enables us to “stand towards” a picture-face as we do towards a real

142

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. human face. He acknowledges that there is a difference in the quality of the “stand”

one takes in each case (the phrase “in some respects” implies as much), yet he also

makes it clear that it is a matter of taking a “stand” in each instance. After all, even the

recognition of mind in flesh and blood humans requires that one adopt a certain

(Dennett would say “intentional”) stance towards them: “My attitude towards him is an

attitude towards a soul” (PI 152). It is a question of attitude whether the face one

“studies” or “reacts to” is in a picture or not, though of course the difference matters a

great deal. Wittgenstein’s talk of “stands” however does not mean that one cannot opt

to play the game: volition is not required in order to recognize the picture as a face. As

if he is concerned that his point might be misunderstood just this way, a few paragraphs

later he muses: “If you say ‘Now it’s a face for me’, we can ask: ‘What change are you

alluding to?”’ (PI 166). Wittgenstein’s querying response to the interlocutor’s remark

here suggests that he does not see how one could see the picture-face as anything other

than a face. This exchange immediately follows a comment that confirms this

interpretation: “One doesn’ttake ‘ ’ what one knows as the cutlery at a meal for cutlery;

any more than one ordinarily tries to move one’s mouth as one eats, or aims at moving

it” (PI 166). Just as one does not consciouslytake a knife for a knife, one does not need

to consciously take a picture-face for a face. A particular “stand” is necessary, but this

is a stand one cannot help but assume, just like Dennett’s intentional stance, which

enables us to see others as mental entities, yet is not the result of any conscious

interpretive decision.

143

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Since aspect-seeing and the intentional stance partake of a certain form, of

“dramatic” pretence, the difference between the perception of a real face andpicture a

of a face - though obviously immense - will ultimately be one of degree, and not of

kind: to again recall Wollheim’s characterization of Gombrich’s position, “seeing

someone’s representation is quitecontinuous with seeing that person face to face - with

all that such a view implies or suggests.” Quandaries, to recall Quine’s discussion of

the projective nature of “dramatic idioms,” will therefore inevitably arise (even ethical

ones: “with all that such a view implies or suggests,” as Wollheim pointed out): doI

injure my friend ifI poke out the eyes in a photograph of his face? Whatever I may

rationally believe, I may not be able to help feeling so. That is, even if I am aware that

a picture of a face is, ontologically speaking, merely a representation, mentality will be

recognized in it nonetheless; and thus, thoughI will not blind my friend and cause him

great pain, I may very well feel that I “injure” the aspects of mind that are “alive” in my

perception of his eyes in the photograph. As with facial perception in general, volition

need play no role here.84 The intentional stance appears not only to be a natural power

of the mind, but an almost irrepressible one as well: whenever given the opportunity,

the mind will adopt an intentional stance towards entities and patterns in our

environment. Indeed, even very simple geometric shapes (such as circles and triangles),

when moving in patterns that can be intentionally interpreted, will spontaneously be

84 It is possible to imagine a person “supressing” a mentalizing response to a representation of a face, but this would be a founded or derivative attitude, non-primary in just the way that Heidegger imagines it as possible for a person to assume a purely theoretical or present-to-hand attitude towards a tool despite the fact that the more fundamental (primordial) relationship of humans (or rather Daseins) to tools is characterized by what he calls readiness-to-hand.

144

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. understoodas intentional agents: a classic study done in 1944 by Fritz Heider and

Mary-Ann Simmel “found that when subjects were asked to watch a silent film in which

geometric shapes moved around, and were then asked to describe what they had just

seen, they tended to anthropomorphize (or ascribe agency to) the geometric shapes.

These subjects used a rich vocabulary of volitional mental-state terms in their accounts,

all of which they attributed to the shapes” (MB 35-6).85 Since people will naturally

adopt an intentional stance even towards images of mere circles and triangles, it is not

so surprising that we spontaneously do the same to pictorial representations that are in

any way recognizable as human faces.

And we are innately sensitized to see faces wherever we can. Indeed, we will

unavoidably recognizeas faces even configurations of objects that clearly arenot “real”

faces at all: consider, for example, the composite fruit and vegetable physiognomies of

Giuseppe Arcimboldo (such as Vertumnus). Our perception of a face inVertumnus is

fully compatible with a consciousness of radical artifice. Indeed, one might say that

Arcimboldo’s paintings are intended to exploit and heighten our sense of “twofoldness.”

However, it would be a mistake to think that the configuration of nonface objects in

Arcimboldo’s painting is somehowinterpreted as a face. Rather, the perception of the

nonface objects and the perception of a face-pattem happenat the same time and in

different regions of the brain: face and object recognition, as cognitive neuroscientific

studies of vision have established, are anatomically distinct and autonomous. Martha

Far ah, in fact, tells us of a patient who, suffering from a profound impairment in his

85 F. Heider and M. Simmel. "An experimental study of apparent behaviour."American Journalof Psychology, 13, 1944.

145

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ability to perceive nonface objects (a condition called “object agnosia”), saw in

Arcimboldo’s Vertumnus only a “face, pure and simple” and did not even realize that

any nonface objects were present (CNV 133-4). There is a part of our brain, in other

words, that is innately sensitized to face-like patterns. Our face perception system will

be automatically activated whenever an appropriate pattern of stimuli is present in our

field of visual perception. As Wittgenstein suggested, we cannot help but see a picture-

face as a face: we do not need to “take” Arcimboldo’s painting as that of a face.

And when we see a face, whether in a picture or in real life, we naturally see it

as expressing mentality as well. Indeed, Gombrich has even formulated this point as a

law of physiognomic representation, which, after the nineteenth-century inventor of the

comic strip (Rodolphe Topffer’s), he has named Topffer’s law: “the proposition that

any configuration which we can interpret as a face, however, badly drawn, willipso

facto have... an expression and individuality.”86 Again, as with Gombrich’s exercise in

The Story o f Art, this “law” is worth trying out for oneself: experiment with a number of

different shapes and configurations. One remarkable psychological fact this exercise

makes clear is how widely one can distort the features, and indeed the whole outline of

the head, before the images pass beyond physiognomic recognizabiiity: after trying this

exercise, one realizes why the various cubist transformations, biomorphic mutations and

figurative distortions of Picasso and Bacon yield images that can still appropriately be

called, and perceptuallyseen as, faces. That perceptual immediacy pertains even under

conditions of minimal or radically altered visual input suggests a problem with those

86 Gombrich, ‘The Mask and the Face,” p. 24.

146

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. popular semiotic accounts of Picasso’s art that would “read” his cubist faces as

conglomerations of purely conventional ideograms or signs for faces (but we will return

to this matter later). The sideways faces - composed of configurations of punctuation

marks - that have become such a regular feature of email correspondence in the last

decade, also attest to the truth of Gombrich’s point, at least about physiognomic

expression if not individuality. Here is a representative sampling of such email faces,

along with the mental-state attributions that are commonly associated with each:

:-) happy :-( sad

:-< upset :-j| angry

:-(0) yelling :-d laughing

;-) winking :-} grinning

:-| apathetic :-o shocked

It is interesting that these “emoticons,” as they are called, are often used to compensate

for what some users of email perceive as the lack of affect (and thus tonal ambiguity) of

the written word, as though only the immediacy of a face might breath life into

electronic script (which, on the one hand, lacks the personal “touch” of handwritten

notes and, on the other, the “tone” of voice messages). The felt need for such quasi

pictorial markers of affect, however, is probably but a sad reflection of the extent to

which a wholly instrumental (information-carrying) conception of language has come to

permeate our relationship to words as such. In any case, whatever pallid kind of “life”

these “emoticons” first managed to convey - after all, such typographical innovations

may have struck those who first stumbled upon them as quite affecting - has now

147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. almost entirely died into convention: but notcompletely so, for so naturally sensitized

are we to faces that their expressions can still sometimes stand outas faces.

Consistent with Topffer’s law, Wittgenstein said of his rudimentary picture-face:

“I can study its expression, can react to it as to the expression of the human face.” The

cognitive systems that underlie and enable the recognition of aspects of mind in a face

do not seem to respect the ontological difference between “real” faces and “pictures” of

faces, though the difference is consciously registered (probably because different and

distinct systems perceive the “faciality” and “representational materiality” of a given

picture-face). If we can recognize mind in a real face, we can recognize mind in a

picture-face as well. Remarkably, as Wittgenstein’s schematic picture-face makes clear,

it takes very little visual detail to activate what one might call our “physiognomic

stance.” Like Wittgenstein’s modest drawing, the psychological studies that

demonstrated an innate preference on the part of newborn infants for faces used highly

rudimentary and schematic face-pattem images. The activation of our innate cognitive

preference for faces does not require illusionistically rendered faces: a circular outline,

two dots for eyes, perhaps a line or a dot for a nose, and a line for a mouth seem good

enough to do the trick. Perhaps the perceptual requirements for neurological activation

are not even this stringent or strict: the suggestion of orality in the “gaping” mouth of a

painted cave, or the blank “stare” of an apple with its top pointed right at us in a

Courbet still life, are often minimally sufficient to trigger our natural and highly

sensitive anthropomorphizing tendencies, though responses will differ widely from

viewer to viewer as debates about what elements are, and are not, anthropomorphic

148

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. within certain paintings is often hotly disputed. Gombrich, intuitively sensitive to just

these aspects of the psychology of perception, expressed understandable wonder at this

fact: “The most astonishing fact about... clues of expression is surely that they may

transform almost any shape into the semblance of a living being. Discover expression

in the staring eye or gaping jaw of a lifeless form, and what might be called ‘Topffer’s

law’ will come into operation - it will not be classed just as a face but will acquire a

definite character and expression, will be endowed with life, with a presence” (A I342).

Indeed, the Hebrew account of creation Genesisin powerfully suggests just how hungry

the human mind naturally is for the appearance of a face, any kind of face:

In the beginning God created the heaven and the earth. And the earth was without form, and void; and darkness was upon the face of the deep. And the spirit of God moved upon the face of the waters. And God said, Let there be light: and there was light. (Genesis 1:1-3, KJV)

Bereshit bara Elohim et hashamayim ve'et ha'arets. Veha'arets hayetah tohu vavohu vechoshech al-peney tehom veruach Elohim merachefet al-peney hamayim. Vayomer Elohim yehi-or vayehi- or. (Genesis 1:1-3)

Prior to the invocation of light, there is already the metaphoric “figure” of a face, even

if that face is of a void, absolutely without form: the “face” of the deep, the “face” of

the waters. At the far reaches of the human imagination, imagining a time even before

the creation of imagination as such, even then and there the mind naturally strains to

o 7 see, and therefore makes out, a face. But are we still surprised that the form such

87 Cf. the beautiful and powerful account of the human hunger for the “outline, the figure, of the person” in the opening pages of chapter one of Susan Stewart’sPoetry and the Fate o f the Senses. This dissertation is, in many ways, simply a further exploration of some of the issues broached by Stewart in that inspiring study.

149

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. imaginings would take would be described by the word “face”(“al-peney,” from

“'panim” for face)? A religious text that can project a face even into the time before

creation is one that understands the face’s fundamental importance: no wonder, then,

that one of the defining traits of its God is that one cannot see his face and live: “You

cannot see My face; for no man shall see Me, and live” (Exodus 33:20; the Torah has

“panim” for “face,” the Septuagint “prosopon”). The Judaic prohibition on graven

images is the ultimate acknowledgement of the face’s value and power.

As minimal as the number of visual clues triggering the recognition of a face

may be, once we recognize an image or object as a face, we will be highly sensitive to

the expressive significance of the spatial relations between its various physiognomic

features. We will inevitably perceive (even if we cannot describe, as Wittgenstein

noticed) the minutest alterations in the configurational relations. The psychologist Egon

Brunswik performed studies that “confirm the extreme sensitivity of our physiognomic

perception to small changes; a shift in the distance of the eye which would perhaps be

unnoticeable in a neutral configuration may radically affect the expression of [a]

mannikin, though how it will affect it is not always easy to predict.”88 As Brunswik

summed up his findings: “Human appearance and especially the face, constitute as tight

a package of innumerable contributing variables as might be found anywhere in

cognitive research.”89 Apropos of this point, Gombrich relates an interesting anecdote

first told by Max Friedlander, “of the bank official who insisted that German bank notes

88 Gombrich, “The Mask and the Face,” p. 25. 89 Egon Brunswik, Perception and the Representative Design of Psychological Experiments, 2nd ed. (Berkeley, 1956), p. 115. Cited in Gombrich, “The Mask and the Face,” p. 25.

150

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. should retain a portrait head in their design. Nothing, he said, was harder for the forger

to imitate than precisely the right expression of these artistically quite insignificant

heads, nor was there a quicker way of discovering a suspect note than simply observing

on the way these faces look at you.” But it is the configuration of the features and not the

quantity of visual details that matters most; caricaturists and cartoonists, as Gombrich

has pointed oat, have long recognized and exploited this fact of visual perception.

Facial perception, after all, is simply a form ofpattern recognition, and a pattern is a

pattern as long as it has the minimal number of elements arrayed in the correct

configuration. If not, life would have been much more difficult for cartoonists like

Charles Schulz and James Thurber! It is because so very little is required to activate our

facial recognition and perception systems that the minimally rendered faces of the

characters in Schulz’s Peanuts strips or Thurber’s comics are so readily recognizableas

expressive of mental states (even if, as with all facial expressions, our mentalistic

attributions may end up being ambiguous or uncertain).

Anytime we see an appropriate “physiognomic” pattern, we see, to some degree

or other, mentality as well. Thus, to recall Gombrich’s point about anthropomorphosis,

we see in such shapes (inanimate as they may really be) “life” and “presence” too:

“discover expression in the staring eye or gaping jaw of a lifeless form... [and] it will be

endowed with life, with a presence.” The perception of aspects of mentality in the

representation of faces is precisely what Wollheim’s misreading of section xi of the

Investigations blinded him to: his account of seeing-in explains how we may see a face

90 Gombrich, Art and Illusion, p. 334.

151

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in a picture, but not how we inevitably see that face (in the picture)as afraid,

thoughtful, pensive, joyful, enraged. Seeing-in, on its own, does not adequately explain

the recognition of mind which is such a crucial part of representational seeing. Of

course, Wollheim does not dispute that we readily see thoughts and feelings in the

people portrayed in paintings (his powerful interpretations of various works make

liberal use of mental-state attributions): the point is simply that his misreading of

Wittgenstein’s notion of seeing-as precludes a satisfactory theoretical account of this

obvious phenomenological fact. Wollheim’s position unfortunately militated against

any acknowledgement at all of the fact that pictures can sometimes be seenas persons

(recall his critique of Gombrich on just this score). On the other hand, Gombrich, more

intuitively sensitive as he was to this aspect of sight, never wavered from his sense that

the perception of life and presence was a common part of our experience of

physiognomic representations, even if he was not able, in the end, to produce a cogent,

plausible account of how this was possible. As his use of the powerful term “presence”

indicates, he was not embarrassed to admit that we can see imagesas mental beings. As

I suggested earlier, I think that is why he was able to produce such powerful and

insightful work on the history and workings of physiognomic expressiveness in visual

art.

If neither Gombrich nor Wollheim managed to account for the perception of

mentality in pictorial representations, however, it is important to remember that they

both consistently asserted that the cognitive and perceptual capabilities underlying our

ability to understand pictorial representations were likely biological in nature. As

152

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Gombrich put it: “our reaction to faces and physiognomic expression may not be wholly

due to learning, and that the mental set which makes us read faces into blots, rocks, or

wallpapers may be biologically conditioned” (AI 341-2). Wollheim, inPainting as an

An, likewise said that “seeing-in appears to be biologically grounded. It is an innate

capacity, though, as with all innate capacities, it requires an environment sufficiently

congenial and sufficiently stimulating, in which to mature. A baby a few days old will

respond to the drawing of a face: fleetingly, of course, but the same goes for all its

responses to the external world” (PA 54). As Wollheim’s remark about the

responsiveness of a baby to the picture of a face at the age of a few days suggests, he

did not have access to the kind of empirical information we are lucky to have in such

abundance today (recall that babies with a median age of but nine minutes show

preferential attentiveness to visual face-pattems; and that a baby less than two days old

will show an ability not only to discriminate between different facial expressions but

also to imitate them; cf. chapter two). This is just to note that Gombrich and Wollheim

both ultimately grounded their theories on an empirical and scientific foundation, and

thus implicitly acknowledged that their claims were fallible in the face of new findings.

Similarly to the way Quine regarded philosophy in general, Gombrich especially, but

even Wollheim too, seem to have held the philosophy of art as fundamentally

continuous (though not, of course, identical) with the natural sciences. Philosophy of

mind, in other words, is notabsolutely distinguishable from empirical psychology, nor

is the philosophy of art. There is, in other words, no whollya priori aesthetics: how

could there be since judgments about art, as the term “aesthetics” itself implies, are

153

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ultimately rooted in sense perception? Once we no longer accept the force of

transcendental arguments, the perception of representations will naturally be understood

as a conceptual field to be illuminated as much by experiments in the laboratory as by

the introspections of the lover of art. If we are now able to understand better the

workings of certain parts of the brain and mind that allow us to improve on the

monumental and seminal accomplishments of thinkers such as Gombrich and

Wollheim, that is surely only due to the fact that we simply enjoy an abundance of

empirical information that simply was unavailable to them less than two decades ago.

Of course, this does not mean that everything about perception will be reducible to

scientific explanations: even Quine held that the “humanly indispensable” intentional

idioms cannot be reformulated in the universal language of the natural sciences. Just so,

aspects, which are such an important contribution of Wittgenstein’s to the fields of

psychology and perception together, are not reducible to the outcome of physical

processes (the dynamics of mental processes, as Davidson pointed out, have “no echo”

in physical theory). Nonetheless, a kind of fallibility that the transcendentally minded

Kant and Hegel would have found incomprehensible is now intrinsic to aesthetics.

Considering how relatively new the synthesis of cognitive neuroscience is, along with

developmental and evolutionary psychology, all theories of art that attempt to

accommodate their scientific findings will be quite unstable for the foreseeable future.

It may very well turn out that much of the scientific findings about cognition and

perception that are made to bear such great explanatory weight in projects such as this

will turn out to be falsified by new scientific findings, though the sheer amount and

154

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. theoretical convergence of recent years (especially concerning facial perception)

suggest that the instability will lie more in the details rather than with the larger

explanatory structures. But who knows? All we can do is formulate the best

explanations we can with the best information we currently have.

In any case, Gombrich’s important point about “life” and “presence” brings us

back to the powerful role of the face in the art of Bacon and Picasso. Noting the

V - / ri ’w . /

sn> ^ ^ wG vf apt h vp j'.vi K-) (fj f e \ / f a / \ " 11 A

Fig. 5 from Rodolphe Topffer,Essai du physiognomie

“astonishing fact” that our mind endows the appearance of a face (even in a formation

of cracks in a wall) with the feeling of life and presence, Gombrich speculates on what

this might tell us about the playful experimentation with facial expressions practiced by

someone like Topffer (fig. 5). In 1845, Topffer published a treatise (Essai du

physiognomie) intended to teach others how to create expressive “cartoon” characters

without drawing from life or bothering even to examine the faces of real people

(activities which Topffer regarded as a waste of time). As Gombrich summarizes his

advice:

... anybody who wants to try should be able to find out the traits in which [a particular facial] expression resides. All he must do is to vary his scrawl systematically. If his first mannikin looks stupid and smug, another with the eyes a little closer to the nose may look less so. By a simple reshuffle of these primitive traits,

155

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. our lonely hermit will find out how these elements and their combinations affect him and us. Thus a little experimentation with noses or mouths will teach us the elementary symptoms, and from here we can proceed, simply by doodling, to create characters. Topffer maintains that the heroes of his [own comic strip] stories thus arose out of his penplays. (A I340)

Because our recognition of facial expressions isinnate, no study from life is necessary.

One need only systematically vary one’s drawings and then gauge one’s psychological

responses to the variations. Gombrich notes that masters at rendering facial expressions

throughout the history of art (Rembrandt and Daumier premier among them) all doubted

the value of life study when it came to the rendering of facial expressions: for facial

expressions change too quickly to be captured by even the quickest draughtsman.

Gombrich writes: “Daumier made fun of Courbet and despised Monet. To him who

never drew from life, the study of ‘plein-air’ effects must have seemed nugatory

compared with the study of human reactions” (AI 355). Those who wish to convey the

appropriate expressions, therefore, are thrown back upon their own innate responses to

the images they have put on paper or canvas.

All of this is fascinating for a number of reasons, but Gombrich is particularly

struck by the fact that the method of playful variation advocated by Topffer’s manual is

unprecedented in the history of art. He asks: “why does this method not develop much

earlier?” Acknowledging that questions of “why” are “dangerous in history,” he

nonetheless proposes an intriguing speculative explanation. Considering the ease with

which Topffer was able to create a sense of mental life simply by means of

systematically varied doodling, Gombrich wonders whether it may not have been the

156

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. “very power” to create a sense of mentality, exploited by Topffer, that also “held it in

check” in the past:

It needs the detachment of an enlightened nineteenth-century humorist to play with the magic of creation, to make up these playful doodles, and to question them for their character and soul as if they were real creatures. To the humble craftsman of earlier periods, the experience may not have been free from half conscious or unconscious fears.... The very laws of proportion and style that held the schemata of beauty together in past centuries may have served this additional aim of preventing too much life from entering the artist’s creations. (AI 342)

Gombrich’s remarks are explicitly presented as speculations of the most adventurous

sort, but our own lengthy exploration of the nature of the mind’s natural responses to

visual images of the human face lends rational plausibility to his ideas. We cannot, of

course, know how people regarded images of the face in the past, but wedo know that

an uncanny, almost magical, liveliness often strikes viewers of picture-faces even in our

own modernity (recall Bacon’s remarks about “injury”), and there is nothing irrational

about this. Who knows how such uncanny feelings of presence and life would have

struck humans living under very different historical conditions, such as those prior to

the development of technologies of mechanical reproduction, which led to the cultural

ubiquity of images of all kinds. Whether or not people ever entirely confused images

with persons (little depends on the unknowable answer to this question), we do know

that it is still very common to experience, along with a full consciousness of the

representational nature of a painting, a feeling that the painting has the presenceof a

person as well. “Her picture smiles down on me from the wall,” one can imagine

someone feeling and saying of a portrait, as does Wittgenstein inInvestigations the (PI

157

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 175). The experience of presence or life in a picture of a person, however, need not be

continuous or uniform. As Wittgenstein notes: “I might say: a picture does not always

live for me while I am seeing it. ‘Her picture smiles down on me from the wall.’ It

need not always do so, whenever my glance lights on it” (PI 175, emphasis in original).

Of course, this passage also makes clear that while the feeling of life need not be

continuous, it definitely is an ordinary part of the way we relate to photographs and

paintings. It depends on the stance one assumes, though the stance will not necessarily

be a matter of volition. To recall our earlier formulation, we not only see personsin

paintings, but also, at times, paintingsas persons. Indeed, it may be that those paintings

which conjure up this feeling of personal presence or life within us also naturally call

forth feelings of aesthetic value as well: an intimation of aura, one might say.

158

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Bacon’s Sensational Portraiture

Picasso’s art evinces a profound, life-long fascination with the human figure as a

whole, but the sheer frequency and variety of his representations of the human head and

face - sometimes reduced to a skull, sometimes transformed into a mask - suggest that

they in particular held a special artistic and personal significance for him. Few other

artists have so relentlessly experimented with different ways of representing and

transforming the human physiognomy. In her 1938 monograph on Picasso, Gertrude

Stein went so far as to claim that nothing else mattered for his art but “the head the face

the human body”: these are all that exist for Picasso,” she flatly declared (PGS 13).

Stein immediately followed this remark with a revealing anecdote meant to indicate the

power of the face’s hold upon Picasso’s imagination: “I remember once we were

walking and we saw a learned man sitting on a bench, before the war a learned man

could be sitting on a bench, and Picasso said, look at that face, it is as old as the world,

all faces are as old as the world” (PGS 13). “And so,” Stein went on to explain,

“Picasso commenced his long struggle to express heads faces and bodies of men and of

women in the composition which is his composition” (PGS 13).

The same might readily be said of Francis Bacon, whose art has been even more

single-minded in its devotion to - or rather obsession with - the human figure,

especially the human face. Bacon’s art, too, was one long struggle to express heads

faces and bodies of men and of women in the composition which was his composition.

It is surely no accident, then, that he regularly expressed an affinity to Picasso in his

159

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. many recorded statements about art. But he did not inherit his interest in the figure

from any other artist; on the contrary, he saw it as absolutely and quite literally natural.

In one of his interviews with David Sylvester, he explained it this way: “I think art is an

obsession with life and after all, as we are human beings, our greatest obsession is with

ourselves. Then possibly with animals, and then with landscapes” (BFB 63). And

tellingly - and from our perspective, unsurprisingly - this obsession “with ourselves”

more often than not took the form of the painted portrait: the appearance of the face,

pushed by means of distortion to the limits of resemblance and recognition, appears to

have held the key to the sensation of “life” that Bacon’s art so relentlessly strove to

“trap” in the image, as he often liked to say in his various interviews. A tension

between the appearance of the figure and the practice of disfiguration lies at the heart of

Bacon’s method: “I’m always hoping to deform people in to appearance,” as he said

about his portraits. This same tension also, of course, lies at the heart of Bacon’s

meaning, though one might more safely say meaningfulness: as anyone who has been

struck by the power of his portraits knows, their “meaning” is difficult to articulate with

words and often is simply experienced at the level of intense feeling. This is, however,

perfectly consistent with Bacon’s intent: to “unlock the valves of feeling and therefore

return the onlooker to life more violently” (IFB 17). Why and how his art of

disfiguration strikes us as it does, therefore, along with the significance of the feelings it

evokes, has posed an enigmatic problem for his critics, and it turns out, even himself:

“What has never yet been analyzed is why this particular way of painting is more

160

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. poignant than illustration” (“illustration” is Bacon’s term for realistic representation).91

In order to understandwhy, we will need a better grasp of the role of the face in the

psychology of visual perception. The face is the key to much of Bacon’s art, as it is for

much of Picasso’s as well. Continuing our study of the face, we will leam much about

their art; likewise, by studying their art, we will leam a great deal more about the face.

% %

On June 30,1972, less than one year before his

death and many years after his enigmatic remarks

about faces to Stein, Picasso would draw a self-

portrait that would betray no slackening in the

psychological intensity of his relationship to the image

of the face (fig. 6). Having finished it, he seems to

have sensed that he had accomplished something

radically new (one notices that faces often play Fig. 6 Pablo Picasso,Self- Portrait, 1972 important roles in “breakthrough” works by Picasso:

recall, for example, thePortrait of Gertrude Stein, Les Demoiselles d ’Avignon, or the

Weeping Woman canvases). The next day, he would tell Pierre Daix: “I did a drawing

yesterday, and I think maybe I touched on something. It’s not like anything I’ve done

before.”92 In the bold, child-like crayon lines that forcefully demarcate the outlines of

91IFB 17. 92 Pierre Daix,Picasso createur (Paris: Seuil, 1987), p. 378. Cited in Kirk Varnedoe, “Picasso’s Self- Portraits” in Picasso and Portraiture, Ed. William Rubin (London: Thames and Hudson, 1996), pp. 110- 179, p. 170.

161

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. most of the face’s features we see a clear stylistic echo of the linear structure of the

1907 Self-Portrait (fig. 7). The three-quarter pose of the earlier work, however, has

given way to a fully frontal posture, obliquely suggesting the striking compositional

convention once reserved for funeral effigies, though Picasso’s unkempt appearance

and lack of clothing in the 1972 drawing works against any hint of formality such a

traditional generic association might import: indeed, we might say that this is a work

that invokes the idea of death without summoning the socially-mediated protections of

rituals or conventions.

There is also a significant difference in the

extent of the outlines of the 1907 and 1972 images.

In the 1907 Self-Portrait, the black outlines firmly

and clearly demarcate the boundary between

Picasso’s head and the surrounding space. As Kirk

Vamedoe has said about the 1907 work, the

organization of the lines conveys a strong sense of Fig. 7 Pablo Picasso,Self- Portrait, 1907 structural coherence: “Picasso conjures his likeness

here from a scaffolding of essentially flat, unmodulated, and evenly weighted angular

slashes and rhythmical arcs, potently locked together by insistent parallelisms,

repetitions, and alignments.”9j While the space implied by this canvas is far from being

illusionistically straightforward (as Vamedoe notes, “the rear of the head is collapsed

into the front... [and] what seems the back of the skull is pulled around to profile

93 Varnedoe, “Picasso’s Self-Portraits,” p. 136.

162

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flatness”), what is remarkable is how the “locked” geometry of its linear patterning

nonetheless projects a decisive sense of structural coherence. There is no confusion

between figure and space within this image. One of the most obvious and striking

formal differences then between the 1972Self-Portrait and that of 1907 is the absence,

in the left side of the later image, of any clearly demarcated outline separating Picasso’s

head from the surrounding space. Absolute absence of outline would be striking

enough, but Picasso leaves clear visual evidence that the shape of the head as a whole

had originally been rendered in a relatively symmetrical (inverted pear-like) fashion,

only to be severely re-worked and partially scrubbed out in the left side of the image.

One can easily discern traces of various once-darker outlines now softened and half

effaced by an apparently unsystematic process of rubbing-out and over-drawing. So

whereas the far right side of the image contains a large, relatively crisply bounded, area

of untouched negative space, the left side harbors a wild jumble of faint marks of

various kinds (thick outlines, thin scribbles, scrubbed-out passages) all ultimately

unified by broad patches of mauve or green drawn in with roughly parallel strokes

evincing great, almost frenzied, manual energy. Contrasted with the 1907Self-Portrait,

one feels that form as such has been put in question in the later work: bold geometry

(arrows, diamonds, and circles) is juxtaposed with an antithetical dissolution of

structure (Vamedoe aptly characterizes the right side of the head as “massively

illogical”).94 That the color field dominating the area of this formal dissolution is done

in a blood-like hue, and that the whole right side of the head appears as if it is suffering

94 Vamedoe, p. 172.

163

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. from a kind of structural leak, strongly suggests that the frenzied area on that side is

meant to represent a kind of symbolic hemorrhage. The presence of a passage of mauve

in the grossly dilated right pupil shows that the subject’s vision is entirely colored by

this terrifying, “late” knowledge.

To be sure, one cannot be certain what Picasso meant when, in his remarks to

Daix on the day after he executed this drawing, he spoke of having “touched

something” in it, that it represented a vision “not like anything” he had “done before.”

Nonetheless, the “massively illogical” dissolution of form in the left side of the image is

surely an important aspect of whatever it is that Picasso felt he had “touched.” Of

course, one might object that a kind of formal dissolution had already been achieved by

Picasso long before in the cubist works he executed during the summer of 1910, in

which the “closed form” and continuous faceting of works such as theSeated Woman of

1909 gave way to the radically open forms of works such as Woman with Mandolin. As

Pepe Karmel observes about the major transformations in Picasso’s style during this

period: “the shift from continuous to discontinuous faceting led Picasso to rethink the

status of the space in the immediate vicinity of the figure. The opening of the closed

form made the border between figure and space into a contested zone” (PIC 72).

Whereas in the Seated Woman of 1909 the facets “lock together to make a continuous

skin,” no such structural closure is evident inWoman with Mandolin (PIC 69). Indeed,

Karmel even notes that the “term ‘faceting’ no longer seems adequate to describe

Picasso’s approach here” (PIC 74). Sculptural solidity has given way to a space of

“free-floating” marks: “Rather than translating the naturalistic contours of the figure

164

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. into a vocabulary of geometric facets, he completely re-imagined them as a series of

free-floating lines and planes - a combination of curves, angles, and rectangular strips”

(PIC 74). This development in Picasso’s style, moreover, led to figures that were

“awkward and incoherent,” an obvious fact critically noted at the time even by

Picasso’s friends and supporters (PIC 75). Such criticisms hit home with Picasso, for,

as Karmel notes, the perceptual coherence of his figures mattered very much to him:

“Picasso’s goal was to produce something recognizable as a figure” (PIC 77). A remark

by Kahnweiler demonstrates at once the kind of uncomprehending response Picasso’s

work of this period received, as well as the fact that that response appears to have

bothered the artist: “Picasso himself often repeated the ludicrous remark made by his

friend, the sculptor Manolo, before one of his figure paintings: ‘What would you say if

your parents were to call for you at the Barcelona station with such faces?”’ (RoC 11,

cited in PIC 76). Kahnweiler himself later wrote of Picasso’s move to the open form

during this period as a “decisive advance,” yet at the time, as Karmel notes, he

“regarded them as ‘unfinished,’ and declined to purchase any of them” (PIC 77). The

dealer Ambroise Vollard, speaking to Picasso’s close friend Max Jacob, even declared:

“He’s gone mad, your friend” (PIC 77).

But if many contemporaries of Picasso, such as Manolo and Vollard, appear to

have found the works of 1910 “massively illogical” in their own way, the openness of

form (even figurative incoherence) these works exhibit needs to be carefully

distinguished from the kind of dissolution of form we see in the 1972Self-Portrait. If

Vollard found Picasso’s work of 1910 incomprehensible - even “mad” - those works

165

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were nonetheless the inevitable consequence of an almost systematic process of

abstraction and experimentation (Karmel’s recent study of this crucial period has done a

great service by meticulously unpacking the logic of its development).95 While the

“irrationality” of these earlier works is accidental, that of the 1972Self-Portrait is

absolutely not. In fact, the reassertion of the closed form in the so-called “neoclassical”

and “biomorphic” works that followed the cubist period indicates that the dissolution of

form as such was never Picasso’s fundamental goal or intent during those earlier years.

Even in the most adventurous biomorphic mutations of the late 1920’s and early 30’s

{Bather with a Ball, for instance), there is rarely any confusion about the spatial limits

and boundaries of the radically altered and unfamiliar organic forms. Indeed, one

suspects that the persistence of formal closure throughout so many different stylistic

transformations belies an ultimately Apollonian bias for structural coherence on

Picasso’s part beneath all of his Dionysian experiments with distortion and

metamorphosis. It is when seen against this life-long background that the “massively

illogical” passage on the left side of the 1972Self-Portrait stands out as novel (“not like

anything I’ve done before”). The large passage of mauve exhibits a wild tactile energy

fundamentally alien to even the most difficult and incoherent of the analytic cubist

works: the latter exude a kind of crystalline, almost mathematical, order when compared

to this late drawing. In fact, Picasso would permit himself, or rather feel himself able,

to exhibit such child-like looseness of technique only in his later years: “it took me a

95 P e p e Karmel, Picasso and the Invention of Cubism (New Haven: Yale Univ. Press, 2003).

166

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lifetime to learn to draw like them,” as he once said of children’s drawings,96 Of

course, I do not intend to suggest that the novelty of this late work is entirely technical

or formal: it is only against the obvious representational meaning of the image that this

irrational passage signifies as powerfully as it does. In other words, it is crucial that the

figure that faces us is in fact recognizable as aface: that is what lends the mauve

passage on the left side its affective force. Indeed, there are other works of Picasso’s

last decade that hint at the looseness of technique we see in this finalSelf-Portrait, yet it

is perhaps the fusing of disorderwith a clearly recognizable physiognomy that provoked

Picasso to see in it something fundamentally new. That Picasso, even in the last year of

his life, would come to this new knowledge by means of the representation of a face, is

consistent with the artistic trajectory of his whole career. As Daix, the friend to whom

Picasso first showed this late drawing, put it: “For Picasso, unlike Braque, the face was

the ultimate test of the validity of pictorial experimentation, and the portrait would

become the ultimate stake” (perhaps this explains why Picasso went around repeating

Manolo’s jibe about the way his cubist faces looked).97

As suggested earlier, only Bacon, among twentieth-century artists, has shown a

concern with the visual representation of the face that matched the obsessive intensity

of Picasso’s own. It is telling, then, to hear Bacon, in connection to his own art, speak

approvingly of the biomorphic works Picasso executed in the late 1920’s: “I think

there’s a whole area there suggested by Picasso, which in a way has been unexplored, of

organic form that relates to the human image but is a complete distortion of it” (BFB

96 Roland Penrose,Picasso: His Life and Work (New York: Harper and Row, 1958), p. 275. 97 Pierre Daix, “Portraiture in Primitivism and Cubism,” inPicasso and Portraiture, pp. 254-295, p. 257.

167

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8).98 Reading this statement, made during one of Bacon.’s interviews with David

Sylvester, Milan Kundera incisively notes: “With this very precise remark, [Bacon]

QQ defines the realm whose exploration is actually his alone” (BPSP 10). But if,

following the “suggestion” offered by Picasso’s biomorphic images, Bacon came to

explore a “realm” of figurative distortion all his own, the dimensions of this realm were

never quite as expansive as that first mapped out by Picasso himself. If Bacon’s

relentless deformations of the human face reach a ferocious intensity never attained by

Picasso’s ultimately more controlled and controlling imagination (only the 1972Self-

Portrait comes at all close), that violent power is achieved (perhaps is even enabled) by

a single-mindedness of vision that precluded the range of formal experimentation

evident in Picasso’s own widely varying treatments of the face. Ultimately, this

difference of style bespeaks a fundamental difference of vision: both Picasso and Bacon

found the ultimate test of their artistic skills in the figuration of the face, yet for very

different (though not ultimately incompatible) reasons. Picasso appears to have been

fascinated by the extent to which the “form” of the face could lend coherence and

structure to a world ever threatened by chaos (hinted at by the mauve that swirls in the

dilated right eye of the 1972 drawing), whereas Bacon appears to have marveled at the

way in which a world that ultimatelywas nothing but chaos could ever - solely by

“chance,” as he liked to say - give rise to the absolutely contingent form of the

individual human face. Rather than the biomorphic works of the late 1920’s that Bacon

98 Interviews with Francis Bacon:1962-1979, new and enlarged ed., David Sylvester (London: Thames and Hudson, 1980), p. 8. References to this text will be given parenthetically with the abbreviation IFB. 99 Bacon: Portraits andSelf-Portraits, Ed. France Borel, intro. Milan Kundera (London: Thames and Hudson, 1996). References to this text will be given parenthetically with the abbreviation BPSP.

168

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. approvingly points out to Sylvester, it seems to me that Picasso’s June 30, 1972Self-

Portrait, with its “massively illogical” dissolution of structural coherence, marks the

closest point of proximity between the two artists. Of course Bacon was already far

along on his own path by then (theThree Studies for Figures at the Base o f a

Crucifixion were painted in 1944; fig. 8).

Fig. 8 Francis Bacon,Three Studies for Figures atthe Base of a Crucifixion, 1944

What is remarkable, though I think far from accidental, is that both artists found in the

image of the face a crucial and fixed reference point for their far-reaching artistic

explorations. Though each would transform the facealmost beyond recognition - and

indeed, sometimes they did inadvertently step across that cognitive and perceptual

threshold - each appears to have held dearly to, and put great stock in, the importance

of visual recognition, resemblance and even identity. Significantly, it seems to have

mattered yery much to them that a face of theirs be recognizable as a face, and often

even as the face of aparticular person: no other artists, since the invention of the

photograph, have done more to expand the possibilities of the painted portrait. As

169

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bacon put it quite clearly while speaking to David Sylvester about two portraits he

painted of his friend Michel Leiris, “I really wanted these portraits of Michel to look

like him: there’s no point in doing a portrait of somebody if you’re not going to make it

look like him” (IFB 146). The point of their distortions was never to gobeyond the

human face, but rather to more fully explore the strange, even auratic, aspects of its

appearance.

* v *

Michel Leiris, a good friend of Bacon’s, emphasized the powerful experience of

“presence” he consistently felt when looking at his paintings, in one of the best essays

written on Bacon’s art. Here is how Leiris put it:

As if the picture had its own life, and constituted a new reality instead of being a mere simulacrum... that feature in a Bacon canvas which is immediately apprehended and asserts itself unequivocally and independently of any sense of agreement or disagreement... is the kind ofreal presence to which his figures attain, even though this presence has no connection at all with any kind of theology. Through the agency of the figures, the spectator who approaches them with no preconceived ideas, gains direct access to an order of flesh-and-blood reality not unlike the paroxysmal experience provided in everyday life by the physical act of love. And this presence is graced with a wild ambiguity, an alluring iridescence, which makes it a sensuous delight, but one so intense that, despite the attractiveness of its painterly vehicle, to some people, repelled perhaps by its searing impact, it can appear wholly abhorrent.100

Bacon frequently spoke of trying to trap what he liked to call the fact of life in his

paintings (“fact” and “life,” both terms of approbation in Bacon’s theoretical lexicon,

appear to be synonyms, though little else is self-evident about their meanings): “as an

100 Michel Leiris, “Francis Bacon, full face and in profile,” trans. John Weightman,Francis in Bacon (New York: Rizzoli International Publications, 1987), p. 6.

170

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. artist you have to, in a sense, set a trap by which you hope to trap this living fact alive”

(IFB 57); “It’s really a question in my case of being able to set a trap with which one

would be able to catch the fact at its most living point” (IFB 54); and, to circle round to

where we began, “I would rather practise the injury in private by which I think I can

record the fact of them more clearly” (IFB 41).

The “real presence” that Leiris notes, and which Bacon himself calls by various

different names (such as fact, life, sensation), is profoundly (and I hope now clearly)

related to the simple fact that Bacon is fundamentally a painterfaces of andfigures.

That, I would propose, is one reason that “appearance” is also such an important

theoretical ideaand practical goal for Bacon: “What I want to do is distort the thing far

beyond the appearance, but in the distortion to bring it back to a recording of the

appearance” (IFB 40). While by realistic, or what he calls “illustrational,” standards, he

may be seen to radically distort the images he paints, from his own point of view, he

nevertheless brings thingsback to a recognizable appearance. It’s worth noting that

while the extent to which Bacon’s canvases actually manage to remain recognizable

likenesses of their models cannot be determined by “objective” criteria, their

achievement of resemblance is attested to by a number of different critics. As Kundera

remarks, not without much wonder at the achievement of identity under conditions of

radical disfiguration:

Looking at Bacon’s portraits, I am amazed that, despite their ‘distortion,’ they all look like their subject. But how can an image look like a subject of which it is consciously, programmatically, a distortion? And yet it does look like the subject; photos of the persons portrayed bear that out; and even if

171

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I did not know those photos, it is clear that in all the series, in all the triptychs, the various deformations of the face resemble one another, so that one recognizes in them some one and the same person. However ‘distorted’, these portraits arefaithful. That is what I find miraculous. (BPSP 11-12; emphasis in original)

As Bacon himself said of his portraits of Leiris: “there’s no point in doing a portrait of

somebody if you’re not going to make it look like him” (IFB 146). This last remark

indicates not only the importance of personal identity to Bacon’s conception of

portraiture, but the more general importance of the perceptual category of the face to its

violent force: faces are (and must be) perceivableas faces, if the images are to have the

effect and power he intends. For we do not need to be able to recognize Leiris in order

to feel the force of the portraits Bacon comments upon (we may not even know who he

is, nor what he looked like); but if we did not recognize Leiris’ face asface, a our

experience of the portrait would be a radically different one. Anyone who claims that

paint as such can exhibit violent force is already speaking metaphorically: one does not

injure stretched canvas anymore than one hurts stones. When Bacon, on the other hand,

speaks of causing “deep injury” to an image, he is being as literal as he can be, since the

image he refers to, having the appearance of a face, possesses the life and presence of a

person.

Bacon neverintentionally distorts a face beyond the point where that image is

recognizable as a face. Of course, in the unpredictable course of actual practice, the

image often would move beyond the point of resemblance nevertheless. Bacon’s

reaction to such moments is significant. Sylvester, at one point, asks about this: “Do

you sometimes find in working on a portrait that, when it’s developing in such a way-

172

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that the paint is very alive and strong, at the same time you’re tending to lose the

likeness of the specific person?” Bacon admits that, “More often than not that does

happen.” But when Sylvester goes on to ask: “Do you feel you want to pull it back by

strengthening some illustrational component?” Bacon replies, tellingly: “No, I think you

just lose it.” Bacon, in other words, is not interested in following his paintingsbeyond

recognition, into some realm of non-figurative abstraction (however, he does admit, in a

comment we will turn to in just a moment, that he does sometimes “pull back” the

image, but this only confirms the central importance of perceptual recognizability to his

art). I think we have to understand his resistance to the abstraction that was dominant in

the art world of his day as indicative of an intuitive sense of the affective power that

was only available to him as long as he remained within the zone of figurative, and

especially physiognomic, resemblance. Not because of what the figure meant, but

because of how the figure could be seen. Seen as, if you will: as a presence, as alive, as

a mind, as a person. At one point in his conversations with Sylvester, after he had been

speaking quite a bit about how much he wanted his paintings to emerge as if by chance

and wholly irrationally, he admits candidly that this is perhaps a theoretical fantasy on

his part. In fact, he acknowledges that he works quite consciously to maintain the

minimal visual (that is, illustrational) clues to foster the recognizability of the face in his

images:

I think that in our previous discussions, when we’ve talked about the possibility of making appearance out of something which was not illustration, I’ve over-talked about it. Because, in spite of theoretically longing for the image to be made up of irrational marks, inevitably illustration has to come into it to make certain

173

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. parts of the head and face which, if one left them out, one would then only be making an abstract design. I think what I very often have talked about has been perhaps a particular theory of mine which is impossible to achieve. Of course one does put in such things as ears and eyes. But then one would like to put them in as irrationally as possible. And the only reason for this irrationality is that, if it does come about, it brings the force of the image over very much more strongly than if one just sat down and illustrated the appearance, which of course millions of art students all over the world can do. But I’m quite prepared to believe that mine is a really far-out and impossible theory. (IFB 126)

Illustration and irrationality, figuration and disfiguration: both sides of these oppositions

are equally necessary to Bacon’s art. His is not an art that moves from reality outwards,

away from illustration, but one that circles and goes back and forth constantly, looking

for a perfect tension between them. This passage also makes clear that he is constantly

checking his responses to the image that unfolds, however haphazardly, however

consciously, on his canvas, remarkably confirming the relevance of Gombrich’s general

argument about the role of schema and correction in art even in the extreme case of

someone like Bacon. Bacon may not be sure what his “schemata” are, but he senses

clearly when there is a need for correction. As with Topffer and Daumier, Bacon’s art

relies on a circuit of perceptual interaction between hand and eyes. It is, after all, in the

response to the image that the power of Bacon’s art lies. Bacon is, as Wollheim would

put it, the painting’sfirst spectator.

By tenaciously holding to appearance, recognition and identity, Bacon

maintained a tight grip on a fundamental source of power for his art in general. In the

following remarks to Sylvester, he searchingly discusses this very constellation of

Issues:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. When I was trying in despair the other day to paint that head of a specific person, I used a very big brush and a great deal of paint and I put it on very, very freely, and I simply didn’t know in the end what I was doing, and suddenly this thing clicked, and became exactly like this image I was trying to record. But not out of any conscious will, nor was it anything to do with illustrational painting. What has never been analyzed is why this particular way of painting is more poignant than illustration. I suppose because it has a life completely of its own. It lives on its own, like the image one’s trying to trap; it lives on its own, and therefore transfers the essence of the image more poignantly. So that the artist may be able to open up or rather, should I say, unlock the valves of feeling and therefore return the onlooker to life more violently. (IFB 17)

I think we can now say that the reason “why this particular way of painting is more

poignant than illustration,” is that Bacon is able to exploit our natural sensitivity to face

patterns while forcing us to register violent distortions to an image we intuitively

recognize as a face. The “violence” of the “sensation” entailed by the simultaneous

registration of both of these “facts” upon our “nervous systems” creates a key element

of the sheer power one experiences when beholding a Bacon canvas. The

“physiognomic logic” that I am arguing underlies Bacon’s art also explains why so little

distortion is ever evident in his rather spare and flat spaces and backgrounds: distortion

of non-human objects would add nothing to the overall effect of his images. Leiris

remarks on this difference of treatment between figure and space as a crucial

characteristic of his canvases: “In the case of Bacon’s pictures - at least those I consider

to be most curiously alive, irrespective of any question of quality - their extreme

intensity seems to me to result from the paradoxical conjunction of two procedures: a

more or less marked distortion of the figures, combined with a fairly naturalistic

175

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. treatment of their surroundings” (FFP 11). Indeed, I suspect that the presence of figural

distortion beyond the “skin” of the human figure and face would permit the viewer to

read the surface of the entire canvas as “painterly” rather than be forced perceptually

and cognitively to confront a violent distortion specifically directed at the flesh of a

recognizably human body. We can now see how Ms varied and contradictory

statements to Sylvester about “injury” and distortion are all true at the same time: he

does injure the people he paints (“the injury I do to them in my work”), in so far as their

images can be seenas the persons they represent; at the same time, he doesnot injure

his friends, since injury is done mereto images; and finally, he does and in fact needs to

injure the image (“who today has been able to record anything that comes across to us

as a fact without causing deep injury to the image?”), for this is the source of much of

the psychological power of his paintings, but his talk of injuring images is meaningful

only because he is familiar with the way they can be seenas persons.

It will be obvious, I hope, why the argument about Bacon that I have been

developing must be opposed to Gilles Deleuze’s influential interpretation, which

perversely refuses to acknowledge the primacy of the face for Bacon’s art. Importing

the ideas about “faciality” that he and Felix Guattari developedA in Thousand Plateaus

(discussed above, in chapter one; Deleuze’s text on Bacon is the first he wrote after his

collaboration with Guattari), he reads Bacon’s portraits as a painterly manifestation of a

process they called in that text the “dismantling of the face”:

As a portraitist, Bacon is a painter of heads, not faces, and there is a great difference between the two. For the face is a structured, spatial organization that conceals the head, whereas the head is

176

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dependent on the body, even if it is the point of the body, its culmination. It is not that the head lacks spirit; but it is a spirit in bodily form, a corporeal and vital breath, an animal spirit. It is the animal spirit of man: a pig-spirit, a buffalo spirit, a dog-spirit, a bat-spirit... Bacon thus pursues a very peculiar project as a portrait painter: to dismantle the face, to rediscover the head or make it emerge from beneath the face. (FBLS 19; ellipsis and emphasis in original)

As is the case here, the original discussion of “faciality” inA Thousand Plateaus made

little sense of the idea that a face might (indeed should) be dismantled, let alone what

the quasi-technical term “head” was supposed to refer to (this much is clear: the “head”

is what remains once the social construct of the “face” is dismantled, but as I noted in

chapter one, their theory of “faciality” is premised on the mistaken notion that the

central role of the face in human forms of life is socially determined). So it is difficult

to evaluate the claims that Deleuze puts forward here. But it is obvious that any reading

of Bacon that does not preserve a fundamental role for the humanface cannot begin to

account for the peculiar presence and force his paintings have been recognized to have:

Leiris and Kundera, as we have already noted, do not make this mistake and are quite

clear that the “resemblance” (Leiris) and “faithfulness” (Kundera) of his faces are

essential to the power of Bacon’s portraits.

Moreover, Deleuze’s failure to acknowledge the role of the face leaves him

unable to provide a compelling explanation of Bacon’s use of the term “sensation,”

which, as the subtitle of Deleuze’s text makes clear (“The Logic of Sensation”), is the

conceptual key to his entire reading of Bacon’s works. As I have argued, one of the

primary sources of the violent sensation of life Bacon hopes to provoke in the “nervous

177

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. system” of his viewers is the tension between figuration and disfiguration precariously

balanced in the way he renders his physiognomies. But because Deleuze does not grasp

the critical significance of the perceptual recognizability of Bacon’s faces (against

which any “irrationality” could be perceived), he is forced to look elsewhere for the

source of “sensation.” That is, in his eagerness to regard Bacon as an artist who is

concerned primarily to “rediscover the head or make it emerge from beneath the face,”

he fails to respect how important the face (as a face) is to whatever power Bacon’s art

possesses. He argues instead, in a cosmological vein, that “sensation” refers to a pre-

reflective life of the senses that is linked to a universal, cosmic “Rhythm” that runs

through all existence. (The source of the Rhythm itself he traces to a primordial chaos:

“We can seek the unity of rhythm only at the point where rhythm itself plunges into

chaos, into the night, at the point where the differences of level are perpetually and

violently mixed”; FBLS 39). According to Deleuze, the sensory experience of any

particular sense modality (such as “visual sensation” in particular)

is in direct contact with a vital power that exceeds every domain and traverses them all. This power is Rhythm, which is more profound than vision, hearing, etc. Rhythm appears as music when it invests the auditory level, and as painting when it invests the visual level. This is a “logic of the senses,” as Cezanne said, which is neither rational nor cerebral. What is ultimate is thus the relation between sensation and rhythm, which places in each sensation the levels and domains through which it passes. This rhythm runs through a painting just as it runs through a piece of music. (FBLS 37)

I admit that I find these ideas quite beautiful, but I am not sure what they have to do

with Bacon’s art. It follows from Deleuze’s account here that “sensation” is not to be

178

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. understood (as I have been claiming) as the result of how weperceive Bacon’s distorted

human figures, but rather the distortions themselves are to be regarded as visual traces

of the rhythmic force of sensations that runs through the bodies on the canvas: “In short,

it is not movement that explains the levels of sensation, it is the levels of sensation that

explain what remains of movement..,. [IJn the end, It is a movement ‘in-place,’ a spasm,

which reveals a completely different characteristic of Bacon:the action o f invisible

forces on the body (hence the bodily deformations which are due to this more profound

cause)” (FBLS 36). But this leaves unexplained whywe - the beholders of Bacon’s

paintings - would so intensely experience the sensations that, according to Deleuze’s

basically formalist account, simply inform or deform the visual dispositions of the

bodies on his canvases. If one accepted Deleuze’s cosmology, one might argue that

because the rhythmic patterns and forces underlying sensations run through the whole

of the universe, the viewer of the sensations represented in a Bacon canvas (visible in

his deformed bodies) would quite naturally be sympathetically affected (I am not

certain, but I believe Deleuze implies as much though he nowhere makes it explicit).

One would have to accept his cosmology, however, for this to seem a compelling

answer.101 But however it measures up as cosmology, I would argue that it falls short as

101 It appears that Deleuze’s cosmology ultimately derives from his reading of Nietzsche’s cosmological doctrine of the “will to power.” For an incisive evaluation of the role of this doctrine within Nietzsche’s philosophy as a whole, see Maudemarie Clark’s Nietzsche on Truth and Philosophy (Cambridge: Cambridge University Press, 1991). Clark argues quite compellingly that Nietzsche did not regard the cosmological theory of the will-to-power as true (she also argues, forcefully, that Nietzsche never abandoned the concept of truth itself, against the prevailing tendency, by both his critics and supporters, to read him this very way).

179

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. an account of Bacon’s art, for the simple reason that it is blind to Bacon’s quite obvious

obsession with the face, as a face.

Finally, Deleuze’s account disregards the fact, noted earlier, that Bacon

constantly gauges his own perceptual and psychological responses to the images he

paints as they emerge on his canvases: perceptual interaction between image and

beholder is not something that can be left extrinsic to any account of the genesis or

meaning of his works. This means, as Bacon candidly admitted, that “illustrational”

elements may be introduced so as to maintain the recognizability of the face: “Of course

one does put in such things as ears and eyes.” It is, however, his ongoing perceptual

response that will determine when and what kinds of alterations will be introduced (and,

of course, this does not entail conscious control, though it does not preclude volition

either: appearance is not inconsistent with chance, an important point which we will

discuss later at greater length). Such perceptual interaction, and the resulting beholder’s

“sensation” that is the measure of the perception’s intensity, are internal to Bacon’s

creative process. In the following remarks, Bacon characterizes this process as a

continuous and relentless form of self-criticism:

You see, one has an intention, but what really happens comes about in working - that’s the reason it’s so hard to talk about it - it actually does come about in the working. And the way it works is really by the things that happen. In working you are really following this kind of cloud of sensation in yourself, but you don’t know what it really is. And it’s called instinct. And one’s instinct, whether right or wrong, fixes on certain things that have happened in that activity of applying the paint to the canvas. I think an awful lot of creation is made out of, also, the self- criticism of an artist, and very often I think probably what makes

180

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. one artist seem better than another Is that his critical sense is more acute. (IFB 149)

Notice that he locates the “cloud of sensation” referred to in these remarks inhimself (It

is a kind of critical “instinct”), reacting to and guiding the development of the image.

Whenever Bacon successfully traps sensations in his images, whatever Is captured is the

result of a complex and continuous perceptual interaction between the viewer (first of

all, Bacon himself) and the organically emerging Image seen, at each moment, as a

distorted though recognizable part of a particular person’s body, often his or her face.

We might say, then, that sensation is not what deforms Bacon’s bodies (like an

“invisible force” pressing down upon them, as Deleuze imagines) but something we

naturally experience (in our own “cloud of sensation”) when we perceive bodies so

deformed. The truth is that Bacon’s “sensations” are probably neither really in the

canvas nor in the viewer, but always a measure (in the “nervous system”) of how the

canvas is perceived by the viewer (whether that is Bacon or us). So understood, it

would not be wrong to then see sensations as either in the image or in the viewer (as

Bacon sometimes does), but neither of these two localizations would make sense

without their emergence in the perceptual contact between the two (“sensation” has no

metaphysical existence, one might say, which is why I think Bacon uses the fuzzy

image of a “cloud”: it only comes to life when something isseen or experienced a

certain way).

One consequence of Deleuze’s disparagement of the concept of the face is that

he necessarily neglects the importance of identity and personality for Bacon’s art:

181

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. indeed, if Deleuze were right, Bacon should probably be regarded as a Mud of anti

portraitist. Yet Bacon’s paintings themselves, along with many of his published

remarks, make dear that he was deeply concerned with what we might variously call

the individuality, personality or identity of the people he chose to paint. Like Picasso,

Bacon only felt like painting people he knew quite well. “I couldn’t do people I didn’t

know very well,” he said: “I wouldn’t want to. It wouldn’t interest me to try and do

them unless I had seen a lot of them, watched their contours, watched the way they

behaved” (IFB 73-4). During the same conversation in which Bacon made these

remarks, Sylvester also called the “most crucial step in the development o f’ Bacon’s

“subject-matter” that time in the early 1950’s when Bacon “started to do paintings of

particular people [he] knew” (IFB 72). And when asked why he did not turn to this

subject matter sooner, Bacon explained that he simply felt technically inadequate to the

task, implying that the change in subject did not mark a change in interest: he was

always fascinated by his friends and their personalities and would have painted them

sooner if he had felt himself capable enough. And in a different interview, Bacon again

links portraiture to personality. When asked by Sylvester about the “distinct presence

or threat of violence” that “most people” feel when looking at his works, Bacon’s

response confirms not only that he agrees that his works exhibit a kind of interpersonal

“violence,” but also that this violence is intimately tied to the way he sees personality as

such:

When talking about the violence of paint, it’s nothing to do with the violence of war. It’s to do with an attempt to remake the violence of reality itself. And the violence of reality is not only

182

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the simple violence meant when you say that a rose or something is violent, but it’s the violence also of the suggestions within the image itself which can only be conveyed through paint. When I look at you across the table, I don’t only see you but 1 see a whole emanation which has to do with personality and everything else. And to put that over in a painting, as 1 would like to be able to do in a portrait, means that it would appear violent in paint. We nearly always live through screens - a screened existence. And 1 sometimes think, when people say my work looks violent, that perhaps I have from time to time been able to clear away one or two of the veils or screens. (IFB 81-2)

One thing these remarks show is how profoundly traditional Bacon’s conception of

portraiture actually was. Far from turning his back on verisimilitude and likeness as

criteria of aesthetic success, he seems to have seen himself as pushing the medium of

paint towards a greater faithfulness to the personality being represented. Bacon did not

move portraiture beyond personhood, but rather re-imagined the “appearance” of

personality as such.

There is, however, a significant and consequential ambivalence about the idea of

personality that runs through these remarks by Bacon. On the one hand, personality is

an “emanation” that, to speak etymologically, flows out from and floats around the

person’s face (here, that of his interlocutor, Sylvester): “When I look at you across the

table, I don’t only see you but I see a whole emanation.” On this view, personality is

out there, as it were, part of the realm of appearances, rather than something hidden

away and private, like one’s innermost soul (“in interiore homine,” to recall Augustine’s

formulation). On the other hand, however, Bacon also implies that personality is, in

fact, hidden or concealed (we live a “screened existence”) and that the violence people

sense in his paintings may have to do with the fact that he somehow sees through or

183

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. clears away those “'veils or screens.” The “distinct presence or threat of violence” that

Sylvester referred to, according to this view, would be an index of the kind of

psychological violation Bacon practices. As is obvious, these two very different, one

could say opposed, ways of picturing personality, have quite different consequences for

our understanding of Bacon’s art, as they may have had for his own understanding of

his practice as well: one view sees his art as a faithful record of the visible, the other as

an invasive disclosure of the hidden. I will return to the latter possibility after first

taking a moment to consider the surprising complexities of the former.

For the visibility of personality does not entail simplicity of transcription. Art,

for Bacon, is not illustration. We can begin by noting that it is of the very essence of

the visibility of such emanations - ever shifting and transforming - that they be

absolutely mysterious and profoundly difficult to capture. As he says about

“appearance” elsewhere:

The longer you work, the more the mystery deepens of what appearance is, or how can what is called appearance be made in another medium. And it needs a sort of moment of magic to coagulate colour and form so that it gets the equivalent of appearance, the appearance that you see at any moment, because so-called appearance is only riveted for one moment as that appearance. In a second you may blink your eyes or turn your head slightly, and you look again and the appearance has changed. I mean, appearance is like a continuously floating thing. (IFB 118)

The essential transience of the appearance of the human body that Bacon refers to here

is what convinced some masters of physiognomic expressions, such as Daumier, to

dismiss the value of drawing from life. However, appearance is not so difficult to

184

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. capture because it is somehow hidden, but only because it is so fleeting. On the other

hand, the fleetingness of appearance is not simply due to the speed with which the

muscles of the human body or face physically alter. This would make the difficulty

purely contingent, something a faster hand or perhaps a camera might remedy. Indeed,

Bacon regularly used photographs as visual references when he painted, and since his

paintings look little like anything a camera could capture, we can be certain that

physical transience has nothing to do with the “mystery” he invokes. On the contrary,

the difficulty appears to be essential to the nature of appearance as such, for appearance

does not seem to be reducible to “physical” or “literal” appearance at all. “I don’t only

see you but I see a whole emanation which has to do with personality,” he said to

Sylvester, as if there were some kind of disjunction internal to the way Sylvester

appeared to Bacon. Note how, in order to explain why physiognomic expression is so

difficult to capture, we were forced into embarrassing linguistic difficulties: did we not,

in effect, just say that appearance is not reducible to appearance? We are at the

borderlands of sense here. Perhaps this explains why Bacon resorts, somewhat

tentatively, to characterizing appearance as “like a continuously floating thing.” This

formulation, along with his reference to an “emanation,” suggests some kind of gaseous

formation or cloud (recall the “cloud of sensation”), perhaps shot through with

variously colored light, ever shifting, yet fully visible and, unsurprisingly, almost

impossible to “trap” (it takes a “moment of magic”). Imagined this way, it is interesting

to recall that Bacon spoke approvingly of Monet’s sunsets, even expressing a desire to

paint a human mouth in the same way: “I like, you may say, the glitter and colour that

185

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. comes from the mouth, and I’ve always hoped in a sense to be able to paint the mouth

like Monet painted a sunset” (IFB 50).

It would be easy to take Bacon’s way of characterizing appearances (as

“emanations” and “continuously floating things”) as indicative of some sort of occult or

spiritualistic views on his part, like those held by so many great modem painters before

him (Kandinsky and Mondrian, for example). At least this would free us from the

contradiction we found ourselves cornered into: understood spiritualistically, Bacon’s

appearances would not be reducible tophysical appearances because the “appearances”

Bacon sees are of immaterial realities (psychic “auras,” perhaps). That is, Baconsees

personality, but with a “sight” that is spiritual rather than mundane. But this reading

creates more difficulties than it resolves. Bacon, for one thing, despised all modem

“mystical” ideas. In response to a suggestion by Sylvester that one can be said to enter

a “trance-like state” while engaged in painting, Bacon’s response is telling: “I don’t like

using the word trance-like nowadays, because it comes too near to modem mysticism,

which I hate” (IFB 96). He also called himself a “non-believer” who professed a

radically disenchanted view of existence: “I think of life as meaningless; but we give it

meaning during our own existence. We create certain attitudes which give it a meaning

while we exist, though they in themselves are meaningless, really” (IFB 23, 133). He

also held a thoroughly naturalistic conception of the place of humans in the animal

world: nothing but random contingency differentiates us from the meat we see at the

butchers, he believed. Humans, after all, are but animals of another kind: “Well, of

course, we are meat, we are potential carcasses. If I go into a butcher’s shop I always

186

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. think it’s surprising that I wasn’t there instead of the animal” (IFB 46). Given the tenor

of these remarks, it is hard to imagine that Bacon would intend anything remotely

supernatural or spiritual in his remarks about “emanations” (or “continuously floating

things”). So if it is clear that the “emanation” that Bacon saw around Sylvester is not

reducible to the physical appearance of Sylvester’s face and body, it is just as dear that

it cannot be anything spiritual or mystical either. Perhaps it would help if we likened

Bacon’s notion of “appearance” to Wittgenstein’s theory of aspects, for aspects too are

neither purely physical nor ethereally spiritual. And like Bacon’s appearances, aspects

are what Wittgenstein said wesee when we see a face as expressive of a person’s

thoughts and feelings: the affinities between the ideas of Wittgenstein and Bacon are

striking. Perhaps this is unsurprising when we consider that few people have spent as

much time and energy reflecting attentively upon the appearance of the human face as

did they (there is no reason, I should note, to suspect that Bacon read or was aware of

Wittgenstein’s discussion of aspects: I think each came to their own views

independently, simply by means of intense observation and thought). That does not

mean, however, that their views are identical: that there are real differences in their

physiognomic views is as undeniable as the unmistakable points of resemblance.

Nothing is to be gained by forcing Bacon’s idiosyncratic ideas into a philosophical

mold, or by using his art as a mere illustration of a Wittgenstelnian theory. This would

add nothing to the cogency of Wittgenstein’s views nor augment the force of Bacon’s

art, the powerful immediacy of which is clear to anyone who cares to look.

Nonetheless, Wittgenstein’s notion of aspects allows us to see that much of what Bacon

187

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. says about appearance is, in fact, rationally cogent (however odd and paradoxical much

of it may sound), and so can show us in what ways they are not “merely” idiosyncratic.

Bacon’s art and theory, should we say, touch upon somethingtrue about faces and their

appearances, and it is in their respective contact with the truth about faces that the work

of Bacon and Wittgenstein meet.

But does that mean that thetruth of facial perception and recognition is

somehow fundamentally violent, as so much of Bacon’s art appears to say? Is there a

kind of epistemological brutality structuring intersubjectivity itself? Consider, for

example, the portrait triptych that he painted of Henrietta Moraes in 1969 (fig. 9).

Fig. 9 Francis Bacon,Three Studies of Henrietta Moraes, 1969

Though criticism often finds the obvious a bit embarrassing to state, is it truly possible

to look at these portrait studies andnot imagine physical violence? Do the various

bloodied faces of Moraes in this triptych series not clearly betray evidence of a terrible

and terrifying brutality that has flayed, mauled and twisted her physiognomy? We have

seen that Bacon does not shrink from the association of violence with his work. There

is no point, I think, saying that his work is notreally violent, that this is somehow a

188

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. naive response, that only “simple" people would see it that way. Not only does Bacon

himself speak of causing “injury” to his portrait subjects (“the injury that I do to them in

my work”), he also acknowledges that people who sense a “distinct presence or threat

of violence” when looking at his portraits may be on to something. He connects this to

his belief that we live “through screens - a screened existence”: “I sometimes think,

when people say my work looks violent, that perhaps I have from time to time been able

to clear away one or two of the veils or screens.” As we noted earlier, Bacon’s

formulation here suggests that personality, far from being visible and part of the realm

of appearances (i.e. emanations), is actually hidden and private, therefore requiring

invasive exploration in order to be successfully uncovered and captured in paint. This

would explain the violence people sense in his work: what they are responding to is the

psychological violation that is intrinsic to his method. And yet, interestingly, this is a

violation which, attempting to get at the personality underneath, only manages to mark

the surface of the body: what psychological content (fear, longing, hope, etc.) is ever

revealed by one of Bacon’s portraits?

One of the people who most forcefully and eloquently responded to Bacon’s

portraits as invasive works of violence was Milan Kundera, who has written one of the

most powerful yet disturbing essays on Bacon’s portraiture. In 1977, just after he had

emigrated to France, Kundera was given the task of writing an essay on Bacon for the

periodicalL ’Arc. He decided to write a brief reading of Bacon’s 1969 portrait triptych

of Moraes, in which he interpreted Bacon’s “brutal” distortions of Moraes’ face as

evidence of the painter’s struggle to “seize hold of her essence, of that diamond hidden

189

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in the depths” (BPSP 10). Paced with the task of writing on Bacon’s triptych, Kundera

significantly resorts to the mode of autobiography and confession, as if, in order to

understand the paintings, one had to subjectively identify with the point of view of the

painter; as if that point of view, and the way it straggled to see or grasp the essence of

another person, was the true “subject” of the paintings. In order to convey the quality of

his response to Bacon’s work, Kundera tellingly says little about the appearance of the

paintings. There is no interpretation of form as such, for the meaning of Bacon’s

images lie not in their forms, but rather in theact of perception and response the

appearance records (the sensation). In order to be as faithful as possible to the

experience of response, Kundera spends much of the essay relating the “scandalous”

and “unconscionable” desire he felt during an encounter with a young woman in Prague

before he had emigrated to France:

It was 1972. I met with a girl in a Prague suburb, in a borrowed apartment. Two days earlier, she had been interrogated by the police about me for an entire day. Now she wanted to meet with me secretly (she feared that she was constantly being followed) to tell me what questions they had asked her and how she had answered them. If they were to interrogate me, my answers should be the same as hers. She was a very young girl who had as yet little experience of the world. The interrogation had disturbed her, and, after three days, the fear was still upsetting her bowels. She was very pale and during our conversation she kept leaving the room to go to the toilet - so that our whole encounter was accompanied by the noise of the water refilling the tank. I had known her for a long time. She was intelligent, spirited, she had fine emotional control, and was always so impeccably dressed that her outfit, just like her behaviour, allowed not a hint of nakedness. And now, suddenly, fear like a great knife had laid her open. She was gaping wide before me like the split carcass of a heifer hanging from a meat hook.

190

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The noise of the water refilling the toilet tank practically never let up, and I suddenly had the urge to rape her. I know what I’m saying: rape her, not make love to her. I didn’t want tenderness from her. I wanted to bring my hand down brutally on her face and in one swift instant take her completely, with all her unbearably arousing contradictions: with her impeccable outfit along with her rebellious guts, her good sense along with her fear, her pride along with her misery. I sensed that all those contradictions harboured her essence: that treasure, that nugget of gold, that diamond hidden in the depths. I wanted to possess her, in one swift moment, with her shit along with her ineffable soul. But I saw those two eyes staring at me, filled with torment (two tormented eyes in a sensible face) and the more tormented those eyes, the more my desire turned absurd, stupid, scandalous, incomprehensible and impossible to carry out. Uncalled-for and unconscionable, that desire was nonetheless real. I cannot disavow it - and when I look at Francis Bacon’s portrait-triptych, it’s as if I recall it. The painter’s gaze comes down on the face like a brutal hand trying to seize hold of her essence, of that diamond hidden in the depths. Of course we are not certain that the depths really do harbour something - but whatever it may be, we each of us have in us that brutal gesture, that hand movement that roughs up another person’s face in the hope of finding, in it and behind it, a thing that is hidden there. (rptd. inBPSP, 9-10)

Kundera’s understanding of Bacon’s artistic method (“the painter’s gaze comes down

on the face like a brutal hand”) implies, indeed presupposes, that violent and invasive

force is the only way to touch and grasp the mind of another: the only way we can truly

know another person, since the other’s self is understood to be a substantive essence

(like a “nugget of gold” or a “diamond”) that lies “hidden in the depths.” Understood in

such a way, how else except by means of force could one imagineseeing let alone

portraying the person beneath his or her “veils or screens,” as Bacon would later put it?

Perhaps this is one reason Bacon was so fond of the word “trap”: regarded this way,

painting might very well feel like a form of hunting. Either way, it is the meat we

191

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. crave. But, of course, the kind of knowledge that is the epistemological goal of this

fundamentally Cartesian picture of mind is impossible to attain (for Cartesians, other

minds will always be aproblem), and Bacon’s aesthetic “success” can therefore be seen

as a reflection of an epistemological failure: only thebody is ever touched. The mind,

as Wittgenstein made clear, is not like a diamond or nugget of gold: it is notthing a at

all to be treasured in absolute privacy (“There are no such things as minds, but people

have mental properties,” as Davidson expressed this point). Bacon’s paintings, as

Kundera’s essay reveals, graphicallyshows the ethical stakes involved when we think

about the nature of mind. Is it private; is it public? Is it visible; is it hidden? Seeing it

as private and hidden, one gropes for the inner self only to find folds of flesh bloodied

and bruised by one’s fists, for there is nothingthere: “Of course we are not certain that

the depths really do harbour something.” Bacon’s paintings can be read, in this light, as

ambivalent cautionary tales, the drama of other minds played out on the surfaces of

their faces: beware of going beneath the surface, his physiognomies seem to say, for

appearances are the only place mind can be seen (whether as aspects or emanations).

This reminds us of Nietzsche’s remarks about surfaces in the Gay Science

(unsurprisingly, Bacon read Nietzsche):

Oh, those Greeks! They knew how to live. What is required for that is to stop courageously at the surface, the fold, the skin, to adore appearance, to believe in forms, tones, words, in the whole Olympus of appearance! Those Greeks were superficial —out of profundity\ And is not this precisely what we are coming back to, we daredevils of the spirit who have climbed the highest and most dangerous peak of present thought and looked around from up there—we who have looked down from there? Are we not,

192

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. precisely in this respect—Greeks? Adorers of forms, of tones, of words? And therefore—artists?

It turns out that Bacon was quite fond of Kundera’s essay. After Bacon died, his

friend Michel Archimbaud even approached Kundera and asked him to write the

introduction to a planned collection of Bacon’s portraits and self-portraits. As Kundera

notes at the beginning of the introduction that he eventually wrote for that volume

(which reprints and expands upon his 1977 essay), Archimbaud assured him that “the

invitation was Bacon’s own wish”: “He reminded me of a short piece of mine published

long ago in the periodicalL'Arc, a piece he said the painter had considered one of the

few in which he could recognize himself’ (BPSP 8). One wonders, even, whether

Bacon might not have had Kundera’s earlier essay in mind when he later spoke of “veils

or screens” in connection to those people who “say my work looks violent.” The

interpretation of his own art in implied by those remarks is certainly consistent with the

essential points of Kundera’s piece. That Archimbaud would entrust the introduction of

Bacon’s collection of portraits and self-portraits to Kundera’s hand shows how genuine

and deeply felt Bacon’s admiration for theL ’Arc essay must have been. However, as

we noted when we first discussed these remarks about “veils or screens,” this is but one

possible view of his art, essentially opposed to that which sees it as a faithful record of

the visible. Of course, Kundera’s reading itself implies this opposite interpretation

when he quite pointedly wonders out loud whether any kind of hidden self exists, for he

does not in fact endorse the idea that portraiture can “pierce” the surface of personathe

in order to uncover the person within. It is, in fact, the very impossibility internal to this

193

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. way of thinking about other minds (“my desire turned absurd, stupid, scandalous,

incomprehensible and impossible to carry out”) that fuels much of the pathos of the

piece, the implication, of course, being that the same pathos (that is, impossibility)

underlies Bacon’s art as well. Like Bacon’s paintings themselves, Kundera wrote the

essay as a mirror of the sensations produced by the perceptual transaction he

experienced while beholding the triptych, and in this lies the significance of the way he

relates the remarks of Archimbaud about Bacon’s feeling for his 1977 essay: “a piece he

said the painter had considered one of the few in which he could recognize himself.”

Recognition (that is, the matching of “clouds of sensation”) is a key to Bacon’s art, and

the mirroring style Kundera chose for his response to Bacon is the most faithful

recognitional response Bacon could have hoped for. No wonder he admired it.

Significantly, this mirroring or matching dynamic already shows a way of perceiving

and representing others that is not invasive.

I do not think, however, that it is a matter of choosing decisively between

competing interpretations. Bacon’s art is powerful not because it is consistent - after

all, it is not philosophy, though it is theoretical - but rather because it dramatizes with

great intensity different ways of approaching and perceiving the human face. In one

respect, it is injurious: flaying, cutting, digging through the flesh. In another respect, it

is not: faithfully recording the transient appearances of emanations of the self as they

continuously float across and about the surface of the face. Both of these possibilities

are simultaneously recorded by Bacon’s portraits, and it is to a great extent the tension

between the two that lends them not only such affective, but also,conceptual power.

194

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. They make us think, not just feel - though we should add that they make us thinkby

making us feel. However, I think these two opposed ways of thinking about the face,

even taken together, do not yet adequately account for the meaning of Bacon’s

physiognomies. There is at least one other important aspect of his art, which has to do

with the meaning of chance in his practice and theory. For though we have referred to

chance repeatedly throughout the course of this discussion, we have not yet fully

accounted for its frequency in Bacon’s remarks about his own practice. What does

chance mean for him? At times, it seems as if it means absolutely everything: “in trying

to do a portrait, my ideal would really be just to pick up a handful of paint and throw it

at the canvas and hope that the portrait was there” (EFB 107). We have already read

enough of his other remarks to know that this is an artistic ideal that Bacon quite clearly

knew was unachievable and unrealistic. As he put it elsewhere, “I’m quite prepared to

believe that mine is a really far-out and impossible theory.” However, his theoretical

fantasies about paintings of absolute accident are nonetheless telling indices of how

strongly he felt about the matter. In response to his acknowledgement that such ideas

are probably “far-out and impossible,” Sylvester says: “Nevertheless, it’s obviously of

great importance for you to think in those terms.” Bacon’s response shows clearly how

much the possibility of a wholly irrational and accidental painting meant to him:

“Certainly. It’s one of the reasons that I go on painting, because it haunts me so much”

(IFB 126).

But just why did it haunt him so? In order to answer this question, I think we

need to return to the radically disenchanted views he held about human life, though

195

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. instead of “disenchanted,” we should probably use more precise terms like “accidental”

or “contingent.” That is to say, for Bacon there is no underlying reason things are as

they are, and not otherwise, which explains his constant existential surprise in the

presence of slaughtered meat: “If I go into a butcher’s shop I always think it’s

surprising that I wasn’t there instead of the animal.” His is a post-Darwinian and

entirely atheistic worldview, absolutely devoid of design, though one that seems (like

the later Nietzsche’s)102 to be based on the acceptance of the truth of the natural

sciences (to the exclusion of all religious belief systems) rather than deriving from a

skeptical commitment to a total absence of truth altogether: hence, his frequent and

positive references to nervous systems, sensations, instincts and drives. Indeed, perhaps

such biological truths are the only ones he fully believed in: “we are bom and we die,

but in between we give this purposeless existence a meaning by our drives” (IFB 134).

But if the drives are real, the meanings they make are not: “I think of life as

meaningless,” he said quite succinctly.103 There is no non-naturalistic teleology

underlying biological life or existence in general. He clearly felt the force of these ideas

quite intensely, and they appear to have permanently colored his adult existence. He

102 On Nietzsche’s naturalistic commitment to the findings of the empirical sciences, see Maudemarie Clark’s Nietzsche on Truth and Philosophy and Christopher Cox,Nietzsche and Interpretation. 103 That Bacon’s use of the word “drives” is not intended in a psychoanalytic sense is made clear by an exchange with Sylvester, in which Bacon strongly resists Sylvester’s attempts to read his portraits psychoanalytically: as signs of ambivalence and contradictory feelings towards his subjects. Here is Sylvester’s suggested interpretation: “do you not think, since you talk about recording different levels of feeling in one image, that, among other things, you may be expressing at one and the same time a love of the person and a hostility towards them - that what you are making may be both a caress and an assault?” Bacon’s response is quite clear and direct: “I think that is too logical. I don’t think that’s how things work. I think it goes to a deeper thing: how do I feel I can make this image more immediately real to myself? That’s all.” Of course, this does not mean that Bacon’s art cannot or should not be interpreted psychoanalytically, it just means that Bacon’s conscious use of the word “drive” should not be taken in that light, since he clearly put no stock in psychoanalytic theories.

196

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. told Sylvester that this Mod of knowledge first came to him one day when, he was

seventeen:

I remember it very, very clearly. I remember looking at a dog- shit on the pavement and I suddenly realized, there it is - this is what life is like. Strangely enough, it tormented me for months, till I came to, as it were, accept that here you are, existing for a second, brushed off like flies on the wall. (IFB 133)

If in the face of its obvious meaninglessness life nonetheless seems to have some sort of

meaning, Bacon saw that as simply due to our own projections: and such fictional

meanings “in themselves are meaningless, really.” That does not mean, however, that

Bacon is unequivocally contemptuous of such fictions. He admits that he has greater

respect (despite his great distaste for all forms of religion) for those who hold

tenaciously to religious beliefs (forms of “total falseness,” as he called them) than those

who, thinking themselves free of morality, “just live a kind of hedonistic and drifting

life” (IFB 134). After all, it takes dedication to hold even to illusions, and “the only

thing that makes anybody interesting is their dedication” (IFB 134). But he admits that

his ideal (or at least “more exciting”) person would be one who could do “totally

without belief,” but nonetheless be “totally dedicated to futility” (IFB 134).

Bacon’s “haunting” theory of the role of accident in art (“one of the reasons that

I go on painting”) derives from this ultimate understanding of the radical contingency of

reality. Crudely put, we might say that on the side of total falseness is illustration, on

the side of truth is chance. Radical contingency, after all, is the only thing undeniably

true about existence. As Gloucester, from Shakespeare’s King Lear, said in lines that

Bacon was fond of quoting: “As flies to wanton boys are we to the gods;i They kill us

197

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. for their sport.” Imposed or willed meaning (that is, illustration) is therefore to be

firmly resisted in favor of an absolute artistic dedication to accident - the practical

aspect, one might say, of a clear-eyed theoretical acceptance of the “futility” of life.

Here we hear Bacon describe how his commitment to accident translated into actual

studio practice: he would, as he explained, “do almost anything to get out of the formula

of making a kind of illustrative image - 1 mean, I just wipe it all over with a rag or use a

brush or rub it with something or anything or throw turpentine and paint and everything

else onto the thing to try to break the willed articulation of the image” (IFB 160). But

what is fascinating is that Bacon did not follow these ideas and practices into total

abstraction, as might initially seem their obvious theoretical and visual endpoint; after

all, what else could be left in the image once illustration is thus banished? On the

contrary, however, he held tenaciously to the appearance of the human figure. Indeed,

not only did he remain committed to the figure as such, but even to the traditional genre

of the portrait, which not only foregrounds figuration but is generically defined by the

resemblance or likeness of the figures portrayed. And as we have seen, Bacon was

surprisingly traditional in his conception of what a portrait properly was. Speaking of

portraiture, we recall that he said “there’s no point in doing a portrait of somebody if

you’re not going to make it look like him” (IFB 146). One might see this as an

inconsistency in his personal philosophy, seeing his devotion to figural resemblance as

at odds with his devotion to accident and chance. For isn’t resemblance necessarily on

the side of illustration? This would be a terribly mistaken inference, however. For

what is so powerful about Bacon’s theory of art is that he sees absolutely no

198

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inconsistency at all in the imagining of a purely accidental formportrait of painting.

Indeed, by devoting himself so often to the painting of portraits, he demanded of

himself that he practice within the zone of this apparent, butonly apparent, paradox:

that individual resemblance could come about wholly by accidental means.

That Bacon’s devotion to figuration and portraiture is not a theoretical

inconsistency, however, requires that we grasp the extent to which Bacon saw realityas

such as wholly accidental. In other words, that particular humans live lives as the

particular humans they are in the first place is the most remarkable piece of

(meaningless) luck one could imagine. “If I go into a butcher’s shop I always think it’s

surprising that I wasn’t there instead of the animal,” is Bacon’s more striking way of

putting the same point. It follows then that a form of portrait painting that devotes itself

to accident should be able to faithfully record the “fact” of this absolute chance, since

the fact of personhood itself is absolutely accidental. What, after all, is an individual

person but a wholly contingent kind of biological entity, which typically exhibit

particular, characteristic, mental properties, sometimes called a personality if they

display a sufficient degree of consistency and coherence over time? If the universe can

give rise, in the total absence of design, to such patterns of organic matter as we are,

then why cannot a totally random painting accurately record their appearance as well?

Hence, his dream of painting a portrait by simply picking up a handful of paint and

throwing it at the canvas. Patterns themselves, Bacon appeared to sense, are absolutely

accidental. Bacon so clearly saw' chance at work everywhere that at times he literally

believed this kind of portraiture should be possible. Indeed, can we say, with certainty,

199

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that it is not? Consider, for a moment, a mathematical analogue to Bacon’s idea.

Imagine flipping a coin one hundred times and writing down the results, denoting heads

by the number 1 and tails by the number 0. We would not be surprised if, after all the

tosses were recorded, we saw the following numerical sequence, which we will call (R):

11000011010110001101111111010001100011011001110111 00011001000010111101110110011111010010100101011110

Consider, on the other hand, the following sequence of one hundred coin tosses, which

we will call (N):

11111111111111111111111111111111111111111111111111 11111111111111111111111111111111111111111111111111

Which sequence of coin tosses is more random than the other? Though (R) intuitively

looks much more random than (N), and though we would be suspicious of anyone who

told us they had honestly recorded one hundred coin tosses that turned out to correspond

to the sequence (N), according to probability theory, neither is more random than the

other. Each sequence of one hundred tosses has exactly the same, very small,

probability of occurrence: [Vi]100, or approximately 1 in lO30.104 In the very way that

this example contradicts our intuitive notions of what randomness “looks like,” Bacon’s

ideas about chance contradict our intuitive notions of what accidental painting will

“look like” as well. Just as the one hundred heads in (N) are just as likely (or rather,

unlikely) as the one hundred tosses recorded in (R), Bacon saw biological patterns (e.g.,

104 This mathematical point and example are taken from an intriguing essay on the relation between randomness and patterns: William A. Dembski, “Randomness by Design”Nous in 25, No. 1 (March 1991), pp. 75-106, p. 80.

200

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a person) as just as likely (or unlikely) as anything else that could possibly come into

existence.

Of course, despite the fact that we cannota priori rule out the possibility that

someone might successfully create the Mud of accidental portraiture Bacons often

fantasized about, such chance likenesses were and are, it goes without saying,highly

unlikely to come about simply by throwing paint at a flat surface. That the fantasy

nonetheless had a strong grip upon Bacon’s imagination (“it haunts me so much,” he

said) meant inevitably that despair would be a constant companion in the studio. And

indeed, few mental states are invoked more often in his interview transcripts than

despair. What is intriguing is the positive, productive uses to which Bacon channeled

this dark feeling. Discussing his desire to feel “freer” while engaged in the act of

painting, at one point he mentions to Sylvester that he uses his “will” to achieve this

goal. Sylvester asks: “The will to lose one’s will?” Bacon’s considered reply - he

decides in the midst of his response that “will” is the wrong word - tells us a great deal

about the role of despair in his art: “Absolutely. The will to make oneself completely

free. Will is the wrong word, because in the end you could call it despair. Because it

really comes out of an absolute feeling of it’s impossible to do these things, so I might

as well just do anything. And out of this anything, one sees what happens” (IFB 13).

The psychological and practical arc described by these remarks is crucial to

understanding Bacon’s art: impossibility leads to despair, which leads to freedom

(“might as well just do anything”), which leads to the unwilled creation of new

painterly possibilities. His ideas about chance, because they are theoretically

201

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Intelligible but practically impossible, put him in a terrible bind: as he himself

acknowledged, “I think what I very often have talked about has been perhaps a

particular theory of mine which is impossible to achieve.” Impossible to achieve, but

not impossible to believe. These remarks suggest how the practical impossibility of

Bacon’s theory is, in a way, the enabling condition for his art as a whole: for his “far out

and impossible theory” forces him inevitably into practical situations of desperation,

which then lead to moments of absolute painterly freedom. Would Bacon even have

been able to experience such moments of freedom at all without this feeling of

impossibility reducing him to despair in the first place?

Bacon’s theory of art, I suggest, was a crucial practical element of his everyday

studio practice (and not, for example, something extrinsic or after the merelyfact:

theory, that is). For this reason, it is important to integrate his remarks into any reading

of his art, not because his statements are privilegedinterpretations of his images, but

because his ideas played an essential practical role in their creation (without them, I

wonder whether he would have painted the paintings he did). By being so frank about

his ideas, it is as though Bacon had invited us into his studio and allowed us to see him

mix his pigments and stretch his canvases. That is, theory was a kind of “tool” right

alongside his paints, brushes and rags. Drink, of course, also played an important

liberatory role in his studio practice during the creation of some early works, such as the

Three Studies for Figures at the Base of a Crucifixion (1944). Speaking of this work,

he told Sylvester: “It was a thing that I did in about a fortnight, when I was in a bad

mood of drinking, and I did it under tremendous hangovers and drink; I sometimes

202

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hardly knew what I was doing,... I think perhaps the drink helped me to be a bit freer”

(IFB 13). However, he came to feel that drink’s artistic utility was ultimately quite

limited: “Sometimes it loosens you, but again I think it also dulls other areas. It leaves

you freer, but on the other hand It dulls your final judgment of what you hold. I don’t

actually believe that drink and drugs help me. They may help other people, but they

don’t really help me” (IFB 54). Alcohol produced (at great cognitive cost) the kind of

freedom which his theory would later, by way of despair, come to supply in a more

practically useable fashion (that is, without loss of critical acuity). In this light, it makes

sense to think of Bacon’s theory as a morereliable substitute for drink: though alcohol

helped him achieve the freedom to paint theCrucifixion triptych (“drink helped me to

be a bit freer”), he told Sylvester that “it’s one of theonly pictures I’ve been able to do

under drink” (IFB 13, emphasis added). The most telling part of his explanation of why

he came to view alcohol as unhelpful is his observation that it “dulls your final

judgment of what you hold”: drink may have helped Bacon paint unwilled images, but

it also dulled his ability torespond to them. And, as I have been arguing, perceptual

response is the key to Bacon’s art - it is in his response to what he has put on a canvas

(whether willed or not) that his “cloud of sensations” (which he also calls “instinct”)

comes to life. Sober despair, unlike sheer intoxication, leads to the creation of images

to which he is then capable of perceptually responding: “At that moment I’m thinking

of nothing but how hopeless and impossible this thing is to achieve. And by making

these marks without knowing how they will behave, suddenly there comes something

which your instinct seizes on as being for a moment the thing which you could begin to

203

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. develop” (IFB 54). And what he sees on the canvas, when “luck” is on his side, is an

unexpected image which hits his “nervous system” with a sensation of presence,life: or

part of the original goal of the impossible theory, it turns out. In the following remarks,

of which we have already quoted a small part, Bacon puts all of this even more clearly:

... you don’t know how the hopelessness in one’s working will make one just take paint and just do almost anything to get out of the formula of making a kind of illustrative image - 1 mean, I just wipe it all over with a rag or use a brush or rub it with something or anything or throw turpentine and paint and everything else onto the thing to try to break the willed articulation of the image, so that the image will grow, as it were, spontaneously and within its own structure, and not my structure. Afterwards, your sense of what you want comes into play, so that you begin to work on the hazard that has been left to you on the canvas. And out of all that, possibly, a more organic image arises than if it was a willed image. (IFB 160)

Illustration is inevitable, for our will is impossible to wholly disengage: it is not as

though Bacon unrealistically imagines that it is possible to consistently paint without

will. He longs for this, but knows it is impossible. But this very knowledge is a source

of great despair, which opens up - in the midst of periods of illustration - moments

beyond willing. And here, in the volatileinteraction between illustration and chance,

new accidental images, with a life of their own, find room to grow: they constitute “the

hazard that has been left to you on the canvas.” And when all goes well, these organic

images produce sensations that approximate those produced by the appearance of the

original subjects, though without being wholly illustrational: “suddenly I have found

that the thing comes nearer to the way that my visual instinct feel about the image I am

trying to trap” (IFB 54). (Here we see again how surprisingly conventional is his notion

204

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of resemblance and portraiture: his portraits would not appear so novel if they were not,

at the same time, so essentially traditional.) Clearly, this process can be repeated over

and over, either until an image successfully traps “the fact at its most living point” or

until it is abandoned and destroyed. Leiris, we will recall, described the uncanny “life”

of Bacon’s images perfectly, when he wrote, “As if the picture had its own life, and

constituted a new reality instead of being a mere simulacrum” (Leiris 6).

Illustration, as Bacon acknowledged, plays an important and inescapable role in

this creative process: in order to maintain the recognizability of the image he will even

introduce ears, eyes, and other “illustrational” elements into his images. Unfortunately,

it appears that he may at times have felt that this was a compromise of his personal

philosophy: his acknowledgement to Sylvester that his theory is “far-out and

impossible” has the ring of a confession (of course, if he was not so emotionally

invested in his theory, it probably would not have worked so well for him: a certain

blindness to its limits may have been the psychological condition of its practical utility).

But far from being an unnecessary and extrinsic element in his artistic practice - despite

whatever Bacon may have believed at times - the role of “illustration” is absolutely

crucial, for the perceptual recognizability of his human figures (especially their faces) is

a fundamental source of the “sensations” he traps in his portraits in the first place.

Bacon’s complicated and unpredictable use of chance distortions was a powerful way of

tuning and altering the sensations we naturally perceive in physiognomic patterns in

order to more powerfully and accurately convey the sensations he perceived when

looking at the subjects portrayed: their personalities, however conceived (as emanations

205

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. or hidden substances). That he married chance and portraiture is essential to the power

and meaning of his art. As I have suggested all along, it is because he represents human

figures and faces that he can so readily speak of his images as being “organic” in the

first place: we naturally respond to such imagesas presences, as Gombrich put it, with a

life and mind of their own. That is to say, Bacon’s intent was simply to create portraits

that could be seenas the persons portrayed.

Following up on this last point, I would like to close this discussion of Bacon by

suggesting that we see him as a fundamentally Romantic artist working under

conditions of modem disbelief, a late and quite altered exemplar, that is, of the creative

powers of the primary Imagination described in Coleridge’sBiographia Literaria. Here

is Coleridge’s well-known definition from chapter 13:

The IMAGINATION, then, I consider either as primary, or secondary. The primary IMAGINATION I hold to be the living power and prime agent of all human perception, and as a repetition in the finite mind of the eternal act of creation in the infinite I AM. The secondary I consider as an echo of the former, co-existing with the conscious will, yet still as identical with the primary in the kind of its agency, and differing only degree,in and in themode of its operation. It dissolves, diffuses, dissipates, in order to recreate; or where this process is rendered impossible, yet still, at all events, it struggles to idealize and to unify. It is essentially vital, even as all objects (as objects) are essentially fixed and dead.

If we replace Coleridge’s references to divinity with accident, I think we are left with a

good picture of one aspect of Bacon’s artistic project: by using chance to create images

that trapped the sensation of life - that is, by creating images that could be seenas

having a life of their own - Bacon participated in the same contingent processes that

206

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. first gave rise to individual human lives in the first place. At those moments of despair

when he let himself “do anything,” Bacon was at one with the only creative force that

exists in the universe: chance (“a repetition in the finite mind of the eternal act of

creation”). Patterns exist: that is a fact, truly, beyond belief. It is as though each act of

painting is, at some level, an act of wonder: wonder at the fact that life exists at all.

And not only life in general, but theseparticular lives: George Dyer, Isabel

Rawsthome, Henrietta Moraes, Michel Leiris, Lucien Freud,himself. Bacon’s portraits

and self-portraits are re-creations in the most literal sense. And by preserving the

aspects that emanate from each face that he saw, he was preserving the presence (for

him) that was each individual person. And when we feel our own “cloud of sensation”

stir when beholding these portraits, that is a bit of their lives come alive in each of us.

Despite Bacon’s disdain for conventional morality, I believe that his pictorial

preservations of the “fact” of each person “at its most living point” is art at its most

ethical (beyond good and evil).

The various ways of reading Bacon’s art that I have presented do not, I know,

easily coexist with one another. But as I suggested earlier, it is not the consistency of

Bacon’s art that matters to anyone who feels the force of his paintings: the force itself is

what is at stake. That various, and perhaps conflicting, theoretical ideas are interwoven

with the affective force of his images does not detract from their artistic value, and I

hope that a reading that is faithful to the multiple theories and perspective that inform

their creation will not be dismissed for that very reason. Indeed, I think it is part of the

strength of his images that they sustain such widely varying (though narrowly

207

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. circumscribed) Interpretations. It is not that anything goes when interpreting Bacon’s

art, but rather that a limited number of ideas and feelings appear to be suggested

simultaneously. What unifies much of his work is not a uniform theory, but the

consistent presence of the human face, and his portraits are one of the greatest

twentieth-century artistic testaments to the importance of the face itself.

208

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5. Face Value

As with the work of the many other figures already surveyed in this dissertation,

Emmanuel Levinas’ reflections on the ethical dimensions of the face-to-face relation

likewise signals the special importance of the human countenance to modem thought/05

The “epiphany of the face qua face,” as Levinas puts it, “opens humanity,” proclaiming

the ethical responsibility of the self for all others (T I194).However, those familiar

with Levinas’ now quite popular and influential work will already know that in his

philosophy the face (visage) is a deeply equivocal term: on the one hand, empirical and

available to sensibility (but only as frozen “caricature” [TI198]), while on the other

hand, transcendent and “neither seen nor touched—for in visual or tactile sensation the

identity of the I envelops the alterity of the object” (TI194). Exceeding the

“enveloping,” and thus violent, grasp of an objectifying sensibility, the face (in the latter

sense) appears as an “epiphany” within the face (in the former): ‘The face resists

possession, resists my powers. In its epiphany, in expression, the sensible, still

graspable, turns into total resistance to the grasp” (TI 197). The transcendent face of

the other appeals to the self and calls it to an absolute responsibility: “Before the hunger

of men responsibility is measured only ‘objectively’; it is irrecusable. The face opens

the primordial discourse whose first word is obligation”(TI 201).More fundamental

than even existence is responsibility, or as Levinas puts it: “preexisting the plane of

105 See especially section 3 (“Exteriority and the Face”) of Levinas’Totality and Infinity: An Essay on Exteriority, trans. Alphonso Lingis (Pittsburgh: Duquesne Univ. Press, 1969). References to this text will be given parenthetically with the abbreviation TI.

209

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ontology is the ethical plane” (TI 201). And so, unavoidable, yet “neither seen nor

touched,” and somehow both before and beyond Being itself, is the transcendent face of

the Levinasian other.

If the deconstructive literary criticism of Paul de Man can be seen as an

exemplary, if rather extreme, instance of modem skepticism about the face and the

personhood it has traditionally stood for, Levinas’ description of the epiphanic

appearance of the face as the very condition of possibility of ethics would seem to

position him diametrically opposite to de Man. And while in a sense this is so

(certainly in spirit),106 Levinas’ positioning of the face outside the field of human

perception is arguably symptomatic of a fundamentally anti-physiognomic orientation

not unlike de Man’s. It is worth asking why Levinas feels it necessary to relinquish the

visibility of the face, indeed to insist on its essential m-visibility. It is clear that his

bifurcated approach to the face is determined by his ambivalent allegiance to, and way

of overcoming, Husserl’s transcendental phenomenology. For Levinas, Husserl’s

model of intersubjectivity is essentially solipsistic, unable to regard the other person as

anything but analter ego constituted by, and in the image of, the self. True alterity is

beyond the reach of the Husserlian “I”. And yet, because Levinas is faithful enough to

Husserlian phenomenology to continue producing descriptions from the perspective of

106 In his remarks about prosopopoeia in his essay on Shelley’s The Triumph o f Life, de Man explicitly rejects the (“naive,” as he declares) notion that the endless tropological process of face-giving and defacement has any “value” for us whatsoever (ethical or otherwise), since “we are its product rather than its agent”: “... to read is to understand, to question, to know, to forget, to erase, to deface, to repeat—that is to say, the endless prosopopoeia by which the dead are made to have a face and a voice which tells the allegory of their demise and allows us to apostrophize them in our turn. No degree of knowledge can ever stop this madness, for it is the madness of words. Whatwould be naive is to believe that this strategy, which is not our strategy as subjects, since we are its product rather than its agent, can be a source of value and has to be celebrated or denounced accordingly” (RoR 122; italics in original).

210

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that “I”, he must regard the alterity of the other person (in the epiphany of her face) as a

radical, or as he sometimes said traumatic, interruption of what Husserl terms the

“primordial sphere” of the transcendental ego, and thus as beyond the reach of

description of any kind. The “view” he thereby achieves of the face is therefore

necessarily vacuous. But is Levinas’ way of approaching the face necessary?

Moreover, by taking a path first determined by Husserl’s transcendental solipsism (if far

beyond its original horizons), does Levinas not risk misdescribing our experience of

other persons and thus the nature of ethical action in the world? Before considering

these questions further, and so as to better appreciate the climate of thought that

motivated Levinas to formulate his radical alternative to Husserl’s phenomenological

epistemology, I think it may be helpful to take a step back in order to reflect upon the

philosophical, and particularly ethical, stakes involved.

Wittgenstein, to use John Ashbery’s memorable phrase, thought of persons (and

their faces) as “visible cores”: mental privacy, according to Wittgenstein’s view, is a

philosophical illusion deriving from a faulty conceptual picture of the nature of thought

and meaning. For Wittgenstein, consciousness in the face of another is directly visible:

as he put it in Zettel, “Look into someone else’s face, and see the consciousness in it,

and a particularshade of consciousness.... Do you look into yourself in order to

recognize the fury in his face? It is there as clearly as in your own breast” (Z 220). In

his later writings, Wittgenstein was self-consciously breaking from a still dominant

Augustinian-Cartesian tradition that pictured the self, to simply reverse Ashbery’s

terminology, as a hidden core; that is, as radically private, precisely in the way that

211

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Augustine describes his own infant mind in the following passage, which we considered

in chapter one, from theConfessions:

Little by little I began to realize where I was and to want to make my wishes known to others, who might satisfy them. But this I could not do, because my wishes were inside me, while other people were outside, and they had no faculty which could penetrate my mind. So I would toss my arms and legs about and make noises, hoping that such few signs as I could make would show my meaning, though they were quite unlike what they were meant to mime.107

It is historically momentous—but perhaps unsurprising given his novel commitment to

the priority of first-person cognition—that Augustine was the first thinker in the western

tradition to formulate the argument from analogy to the existence of other minds. Here

is the way he formulates the argument in his De trinitate :

For what is known so intimately, and so perceives itself to be itself, as that by which also all other things are perceived, that is, the mind itself? For we recognize the movements of bodies also, by which we perceive that others live besides ourselves, from the resemblance of ourselves; since we also so move our body in living as we observe those bodies to be moved. For even when a living body is moved, there is no way opened to our eyes to see the mind, a thing which cannot be seen by the eyes; but we perceive something to be contained in that bulk, such as is contained in ourselves, so as to move in like manner our own bulk, which is the life and the soul. Neither is this, as it were, the property of human foresight and reason, since brute animals also perceive that not only they themselves live, but also other brute animals interchangeably, and the one the other, and that we ourselves do so. Neither do they see our souls, save from the movements of the body, and that immediately and most easily by some natural agreement. Therefore we both know the mind of any one from our own, and believe also from our own of him whom we do not know. For not only do we perceive that there is

10/ St. Augustine, Confessions (Harmondsworth: Penguin, 1961), p. 25.

212

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a mind, but we can also know what a mind is, by reflecting upon our own: for we have a mind. (De trinitate, viii.6.9)

The existence of other minds, according to this argument, can only be established by

means of a process of analogical inference: 1 mustinfer by virtue of the movements of

another’s body, that he or she—Just like me—must have a mind as well. But as

Wittgenstein observed in thePhilosophical Investigations, in reference to the common

idea that it is only by analogy from our own private experiences that we can know what

the word “pain” means, then “must I not say the same of other people too? And how

can I generalize theone case so irresponsibly?” (PI §293). A generalization made on

the basis of one instance to all entities of a certain class is, as Wittgenstein puts it,

logically irresponsible. If we know that others have mental experiences, and we

certainly do (just as we can recognize quite readily that others feel pain), this cannot be

the result of an analogical inference, which would be rationally unjustifiable in any

case.

It is not that the problem of other minds is intractable, but rather that there is no

real problem to begin with. For Wittgenstein, the “problem” of other minds was

anything but: far from being a pressing philosophical question, it was a delusional

“picture” to be therapeutically dispelled. On the other hand, for the Augustinian-

Cartesian tradition (of which Husserl is arguably the most important modem

representatives), there were few more urgent questions than how one could know the

mind of another person (after all, a self-consciously systematic philosophy such as

Husserl’s that had no convincing account of the experience of intersubjectivity could

213

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hardly be considered “complete”). One purely quantitative indication of the importance

of the problem of other minds for Cartesian thinkers such as Husserl is the fact that the

fifth lecture of his Cartesian Meditations, specifically devoted to the explanation of

intersubjectivity, is nearly as long as the other four lectures put together. For those who

accepted the Augustinian concept of mind consolidated in the writings of Descartes, in

other words, the problem of other minds was quite real, perhaps even insoluble.

After all, according to Descartes’ ontology the universe was composed of only

two different substances— res cogitans (thinking matter) andres extensa (extended

matter)—with all traffic between the two necessarily ruled out (except, of course,

through the pineal gland, an implausible “solution” not even Descartes himself seems to

have wholeheartedly believed). If thought itself, according to the Cartesian system, can

have no connection to the extended matter of the body, then there would be no reason to

trust that the behavior (verbal or otherwise) of those around us revealed anything

reliable about the contents of their minds: indeed, there was no reason to believe there

even were other minds. Who was to reassure us that those around us were not, in fact,

automatons? Indeed, Descartes’ thirdMeditation made clear that only reasoned faith in

the existence of a benevolent God was sufficient to settle the extreme form such

epistemological skepticism was bound to take if left theologically unchecked. But even

with the intersubjective world won back by means of faith, radical contingency between

inner and outer would still determine the nature of our access to other minds. As we

noted near the beginning of this study, this is a conceptual situation about which the

214

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. melancholy Hamlet would have shown little surprise: “I have that within whichpasses

show7,” is the way he succinctly expressed this essentially Augustinian idea.

Consider, then, what it would mean for our understanding of intersubjectivity if

minds were private in the very way described by the subjectivist tradition founded by

Augustine and later consolidated by Descartes. This is the situation Levinas felt he had

inherited from his teacher, Husserl, who, at least as far as Levinas knew, had expounded

his fullest treatment of the problem of other minds in theCartesian Meditations.

Levinas would often repeat the fact that it was his disappointment with the results of the

fifth of these lectures, about intersubjectivity, that provoked him to leave the

phenomenological fold (though perhaps more in the manner of a nostos-bound

Odysseus than the wandering Abraham he regarded himself to be). But so as to

understand Levinas’ grasp of the philosophical stakes, it is worth pausing to consider

seriously the conceptual “picture” he had been taught. How, under conditions of

modem skepticism about both God and pineal glands, might the problem of other minds

look? What, in other words, if the self was in fact ahidden core which therefore must

be rendered visible (perhaps even by means of rending) in order to be understood?

• Picturing Privacy: Rilke, Bacon and Coetzee

Unsurprisingly, in the twentieth-century it was various works of literature and

art that expressed most powerfully the tragic pathos of the “theoretical” situation with

which Levinas had to grapple. For example, in a passage from Rainer Maria Rilke’s

215

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. novel, The Notebooks ofMalte Laurids Brigge, the speaker, a Danish writer living in

Paris named Brigge, describes a disturbing encounter that provocatively stages the

possibility that a person, or psychological subject, is indeed an absolutely private

substance and therefore detached from all material reality, including even the

corporeality of the face (staging, in an aesthetic register, the Cartesian problem of other

minds):

Have I said it before? I am learning to see. Yes, I am beginning. It’s still going badly. But I intend to make the most of my time. For example, it never occurred to me before how many faces there are. There are multitudes of people, but there are many more faces, because each person has several of them. There are people who wear the same face for years; naturally it wears out, gets dirty, splits at the seams, stretches like gloves worn during a long journey. They are thrifty, uncomplicated people; they never change it, never even have it cleaned. It’s good enough, they say, and who can convince them of the contrary? Of course, since they have several faces, you might wonder what they do with the other ones. They keep them in storage. Their children will wear them. But sometimes it also happens that their dogs go out wearing them. Any why not? A face is a face. Other people change faces incredibly fast, put on one after another, and wear them out. At first, they think they have an unlimited supply; but when they are barely forty years old they come to their last one. There is, to be sure [natiirlich], something tragic about this. They are not accustomed to taking care of faces; their last one is worn through in a week, has holes in it, is in many places as thin as paper, and then, little by little, the lining shows through, the non-face [Nichtgesicht], and they walk around with that on. But the woman, the woman: she had completely fallen into herself, forward into her hands. It was on the comer of rue Notre-Dame-des-Champs. I began to talk quietly as soon as I saw her. When poor people are thinking, they shouldn’t be disturbed. Perhaps their idea will still occur to them. The street was too empty; its emptiness [Leere] had gotten bored and pulled my steps out from under my feet and clattered around in them, all over the street, as if they were wooden clogs.

216

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The woman sat up, frightened, she pulled out of herself, too quickly, too violently, so that her face was left in her two hands. I could see it lying there: its hollow form. It cost me an indescribable effort to stay with those two hands, not to look at what had been tom out of them. I shuddered to see a face from the inside, but I was much more afraid of that bare flayed head [faloBen wunden Kopf] waiting there, faceless [ohne Gesicht]. (93, trans. Mitchell)

One of the most striking things about this passage is the sudden and sharp tonal shift

that comes at the moment that Brigge recalls the pensive woman at the comer of rue

Notre-Dame-des-Champs. Before the woman is introduced, a disarming levity

characterizes his remarks on faces. He toys playfully with the idea that a face is but a

fashioned thing: it can fit one’s head like a glove, split at the seams, wear thin with use,

and can even be worn by one’s dog - “And why not? A face is a face.” There is no

need to take faces too seriously, he implies; after all, each one of us has many of them,

as an actor might carry around a store of masks in a trunk. Even the suggestion that

there is something “tragic” about those who run too quickly through their finite supply

is qualified by a rather facetious “natiirlich” that deflates the sentiment and works to

maintain the note of supercilious detachment that colors the first half of this passage.

And the sight of those who have used up their supply of faces, thus forced to walk

around with the non-face (Nichtgesicht) showing through, seems to leave Brigge

unperturbed (moreover, the pat internal rhyme between “nicht” and “gesicht” also

works subtly against any inflation of tragic pathos). The tone remains unremittingly

cool, as though Brigge were engaged in a bit of light-hearted cultural critique that did

not threaten to rebound upon himself in any way (the detail of the pet dog walking

217

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. around wearing one of its thrifty owner’s faces is perhaps intended to indicate that the

speaker’s target is something like the Parisian petit bourgeoisie in general). Indeed, the

light tone of the remarks in the first half of this passage makes It easy to read Its

references to faces in a purely allegorical fashion: of course they are not aboutreal

faces, but rather “Identities” orpersonae in an entirely figurative sense.

With the introduction of the woman, however, all this changes. Attention is

suddenly narrowed down from broad generalities onto a single, and singular, person,

who occupies a concrete and specifically named, but at the same time surreal, urban

location. The most salient fact about the woman herself seems to be her immersion in

thought: “she had completely fallen into herself,” as Brigge puts it. Thinking(denken)

is figured as a fall out of the shared world of the social into a deep and absolutely

private interiority. A surreal emptiness (Leere) haunts the street: its cause is left

undetermined, but it is difficult not to feel that it is somehow linked to the figure of the

pensive woman, as though absorption in thought resulted in the evacuation of the

fullness of the world, leaving the street itself “bored,” indicative perhaps of an

insufficiency of human life and interaction. It appears that the woman is meant to stand

parodically for a kind of philosopher, and Brigge seems to have walked into an

agonistic and characteristically modem struggle betweenmind andworld, a struggle

tellingly mapped onto the fleshly contours of a human face.

The contrast between the treatment of the woman and the earlier remarks on

mask-like faces is highlighted by the difference in the way defacement is figured in

each instance. As noted already, the non-face in the earlier part of the passage is

218

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. described with neutral straight-forwardness, the tone suggestive even of a kind of

deadpan and clear-eyed humor (after ail, the joke’s onthem). But the description of the

woman’s loss of face at the end is clearly altogether different, and the way Brigge

characterizes it makes it difficult to read the closing lines in anything but the most

literal way. The interpretive strategy of allegoresis comes to feel profoundly inadequate

to, and almost like a defense against, the corporeal particulars of the moment (we might

think of this as an analogue to the hermeneutic difficulty posed by Bacon’s canvases of

flayed and wounded physiognomies). Perhaps most significant is the fact that Brigge

does not actuallysee what the woman’s head looks like without her face: with an

“indescribable effort” he managesnot to look at her head, though he nonetheless goes

on to describe it as “bare” and “flayed [wunden],” conjuring forth, as if despite himself,

images of exposed viscera and streaming blood. Defacement here is rendered

shockingly literal, and the cool detachment of the earlier part of the passage is

completely lost.

But of what does the conclusion of this passage gives us an oblique glimpse?

(A sight that we must keep in mind is acknowledged to be wholly imaginary, not

grounded in an actual visual perception, even within the terms of the fictional world

described.) It is difficult to say: the ending is profoundly suggestive, affectively

charged, and yet conceptually vague. We are left certain that this flayed head must

mean something, and yet we are left with images of gore obstructing our interpretive

speculations. Despite the fact that the closing lines are explicitly not based on any

perceptual facts, it seems as though the description is intentionally engineered to

219

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. overload the sensory aspects of our imagination with pictures of flesh, blood and,

perhaps most importantly, intimations of bodily pain. For how can we imagine what we

are asked to imagine (violent tearing m d flaying, to use variations on three words

Brigge himself resorts to) without, at the same time, imagining excruciating pain: the

kind of overwhelming pain that can temporarily annihilate our very sense of self and

world? (And isn’t some of the disorientation that we feel in response to this scene due

to our bewilderment as to where the woman’s self or mind in fact resides: in her head or

on her face? And, between these two possibilities, are we not being asked to make an

absolutely impossible judgment, which only feeds back into the feeling of disorientation

that colors the whole conclusion of the passage in the first place?). If the affective tone

of this closing description is so charged, it is surely not simply because Brigge claims to

be “afraid” (though we can certainly sympathize with that response!), but rather

primarily because we are being coerced into picturing a sight that cannot fail to elicit

sensations of pain and aversion that blur the boundaries between ourselves and the

imagined woman. It goes without saying that our only access to the scene is through

the viewpoint of the narrator, Brigge, and so we inevitably experience affective

identification with the fear and disorientation he feels before the strange event that he is

witnessing. And yet, the corporeal intensity of the image that he relates splits, or rather

stretches, our imaginative consciousness to contain (in an indeterminate form: we don’t

know what she looks like or is thinking about) the bodily experience of the faceless

woman as well (and that there is a distinction between body and mind, in the case of the

woman, is something, I think, that wecannot take for granted: in fact, perhaps one point

220

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the passage is to put that distinction into question, if not necessarily to resolve the

issue either way). The surrealism of this passage (e.g., the personified “emptiness”

clattering around in Brigge’s “steps”) does not soften the impact of the reader’s

empathetic response to an imagined body in pain: indeed, the disorientation that the

surrealistic elements of the passage introduce has the opposite effect, rendering the

reader’s imagination susceptible and vulnerable to images that would ordinarily be

patently unbelievable.

But what are we seeing (or not seeing) in the woman’s “bare flayed head”? One

thing the whole passage seems to suggest (though in an equivocal way) is that faces are

only contingently related to persons: faces are like masks that can be removed, worn

through, or even violently tom off - the difference between the last possibility and the

first two, however, is obviously significant and seems to be linked to the absorption in

thought the woman was experiencing at the moment she was startled. She had

completely disengaged from the world, “had completely falleninto herself [in sich],”

and at the jarring sound of clattering steps “pulledout of herself [aus sich], too quickly,

too violently.” The sense of movementin andout of the mind is carefully signaled by

prepositional markers in the description that paint a picture of deep psychic interiority

(a sense of spatial depth that, by contrast, seems utterly lacking in the descriptions of

those who walk around with a non-face earlier in the passage: perhaps their loss of face

is so undisturbing because there is no “inner” self—only absence or nothingness

[Nicht]—lying behind or beneath their socialpersonae). When the thoughtful woman

is startled, she is deep within herself and pulls out violently without making adequate

221

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. adjustments for being in the world: without, that is, donning her facepersona. or

Person and persona are imagined as radically distinct and incommensurable. Her “bare

flayed head,” therefore, seems to show us a gory literalization of what personhood

might look like (though of course we cannot actually see it) if we conceive of the self as

a private inner substance: that is, a hidden core that we might imagine can be rendered

visible only by means of a kind of rending.

But we must be careful to keep in mind that the entire passage is mediated by

the consciousness of the narrator, Brigge. In other words, if we are being presented

with a description in which the woman’s “bare flayed head” literalizes—that is,

incarnates—a Cartesian conception of the self as a private inner substance (a hidden

core), then this is so only from a particular point of view: a point of view that lies

outside the unseen self of the woman, which therefore explains why that self must

necessarily remainunseen. For no matter how thoroughly one may rend the contingent

shells (whether body or face) that exist around a Cartesian self, that self cannot—by

definition—ever be seen or touched by another person. The epistemological distance

between two such selves is infinite and unspannable (for, of course, such selves are not

extended things, and thus lie outside space altogether). In a Cartesian universe, there is

no non-theological resolution to the epistemological problem of other minds.

Of course, that doesn’t mean that we cannot try (or imagine trying)see, to truly

see, another Cartesian self. But Rilke seems to suggest that such imaginings, such

attempts at “seeing,” will inevitably be tainted or contaminated by a kind of

epistemological violence; and, we might add, the fact that the violence in Brigge’s

222

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. description of the woman has no physical agent does not in any way lessen the

atmosphere of violence that suffuses the closing description, for the violent force

resides in the imagination of the scenario itself. Force cannot span the epistemological

chasm between two Cartesian selves, but an attempt by one such self to reach out and

grasp another will very likely, and symptomatically, take the form of force. The

attempt being futile, force will be a measure of, and kind of rage against, that futility

itself.

Rilke’s passage, therefore, can be read as a cautionary tale against such

Cartesian imaginings in the first place. For the descriptive violence that erupts in the

closing of that passage is indicative not of some kind of truth about the self, but rather is

symptomatic of a particular way of thinking about the self. Brigge’s encounter with the

woman suggests that if we think of the self as a hidden core,this (the unseen head of the

woman) is what it might look like to actually see one. But we cannot! And indeed, the

passage seems to say, if given the opportunity, one would avert one’s eyes however

“indescribable” the effort necessary. The point, however, is not that we should or

should not look away, but rather that we ought to reconsider the concept of self that

underwrites such unbearable imaginative alternatives in the first place. That is why it is

so important that the whole of the passage is mediated by Brigge’s consciousness: for it

is the way personhood is envisaged by a particular consciousness (by any

consciousness) that constitutes the true point of the passage. The woman’s “bare flayed

head” can therefore be read as a symptom of, and warning against, a Cartesian

understanding of mind that reifies mentality as such. Do we imagine the self to be a

223

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. hidden core? If so, what are the implications? What are the stakes? Surely, what

Brigge sees (as well as what he does not) gives him good reason to feel “afraid.”

sfc

As I already suggested in an earlier chapter on Francis Bacon’s portraiture,

various aspects of Bacon’s artistic practice can also be seen as a testimony to the fact

that violent force haunts the epistemology of other minds when the person is imagined

as a kind of hidden substance. It is worth briefly recalling in the context of the present

discussion that Milan Kundera’s unsettling essay on Bacon’s 1969 portrait-triptych of

Henrietta Moraes posited a compelling link between mental privacy and aesthetic

violence, when he interpreted Bacon’s distortions of Moraes’ face as evidence of the

painter’s struggle to “seize hold of her essence” (BPSP 10). “The painter’s gaze comes

down on the face,” wrote Kundera in that piece,

like a brutal hand trying to seize hold of her essence, of that diamond hidden in the depths. Of course we are not certain that the depths really do harbour something - but whatever it may be, we each of us have in us that brutal gesture, that hand movement that roughs up another person’s face in the hope of finding, in it and behind it, a thing that is hidden there, (rptd.BPSP, in 9-10)

Kundera’s interpretation of Bacon’s artistic method presupposes that violent and

invasive force is the only way to touch and grasp the mind of another: the only way we

can truly know another person, since the other’s self is understood to be a substantive

essence that lies “hidden in the depths.” Understood in such a way, as we noted earlier,

how else except by means of force could one imagine seeing, let alone portraying, the

person beneath his or her “veils or screens,” as Bacon himself put it? But, of course, the

224

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. kind of knowledge that is the epistemological goal of this fundamentally Cartesian view

of intersubjectivity is impossible to attain (for Cartesians, that is, other minds will

always be problematic), and Bacon’s aesthetic “success” can therefore be seen as a

reflection of an epistemological failure: only thebody is ever touched. The mind, as

Wittgenstein’s later writings make clear, is not like a diamond or a nugget of gold that

might be treasured in absolute privacy: indeed, it is not a “thing” at all (whether

substance, essence or fact). Bacon’s paintings, as Kundera’s reading suggests,

graphically show the ethical stakes involved when we think about the nature of mind. Is

it a hidden thing; is it a thing at all? Is it public; is it private? Seeing it as radically

private, one gropes (like Othello) for the true inner self only to find folds of flesh

bruised by one’s conceptual graspings, for there is nothingthere: “Of course we are not

certain that the depths really do harbour something,” as Kundera himself acknowledged.

Like Rilke’s vignette about the faceless woman, Bacon’s paintings can be read, in this

light, as ambivalent cautionary tales, the drama of other minds played out on the

surfaces of their faces. Beware of going beneath the surface, the distorted likenesses of

Moraes seem to say, for it is only in the realm of appearances that one can see another’s

mind: “the light in other people’s faces,” as Wittgenstein suggestively put it.

* * *

Following on this point, and before finally returning to Levinas, I would like to

look at the role of the face in J.M. Coetzee’s Waiting for the Barbarians, another work

of the modem imagination that, like those of Rilke and Bacon, engages critically with

the Cartesian view of mind that I have been discussing. At one point in the novel, the

225

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Magistrate even experiences a feeling of “dry pity” for torturers such as Colonel loll

that could just as well have been provoked by Kundera’s reading of Bacon: “how

natural a mistake,” the Magistrate thinks to himself, “to believe that you can bum or

tear or hack your way into the secret body of the other!” (43), In the character of M l, a

representative of the fictional Empire’s “Third Bureau” (the division of the “Civil

Guard” charged with intelligence and security), Coetzee presents us with a savage

caricature of the Cartesian theory of mind, transposed into a political register.

Convinced of the privacy of other minds, M l resorts to quite literally invasive means in

order to extract the “truth” that he assumes is hidden deep within his victims. As he

explains his interrogation methods to the Magistrate: “First I get lies, you see - this is

what happens - first lies, then pressure, then more lies, then more pressure, then the

break, then more pressure, then the truth. That is how you get the truth” (5). In other

words, appearances are necessarily associated with deception: as a matter of principle,

prima facie evidence is to be rejected. First lies, then pain, and then only after much

more pain, does one ever arrive at the “truth”: at the “diamond hidden in the depths,” as

Kundera might have put it. For M l, it is axiomatic that truth can only be foundbeneath

a surface that must first be forcefully penetrated; his theory of mind, therefore, has no

room for the possibility that a prisoner might simply tell the truth. Indeed, the

Magistrate contemplates this very possibility with horror, and asks M l about it: ‘“What

if your prisoner is telling the truth,’ I ask, ‘yet finds he is not believed? Is that not a

terrible position? Imagine: to be prepared to yield, to yield, to have nothing more to

yield, to be broken, yet to be pressed to yield more!”’ (5). M l is unmoved, and his only

226

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. response to this hypothetical query is the methodological declaration that deceit

inevitably precedes the disclosure of truth. After hearing loll explain the logic of his

method, however, the Magistrate naturally remains uneasy, and concludes that Joll (like

anyone who uses torture) has mistakenly equated truth with pain: “Pain is truth; all else

is subject to doubt. That is what I bear away from my conversation with Colonel loll”

(5). And indeed, the reader soon learns that the Magistrate is quite right to worry about

the soundness of the Empire’s methods.

Just after this conversation, Joll tortures a young barbarian boy and his uncle,

both of whom had originally ventured near the town simply in order to find a cure for a

skin infection that would not heal on its own: “We were on the road,” the uncle

explained to the Magistrate and Joll, “coming here to see the doctor... He has a sore that

does not get better” (3-4). Disregarding the obvious truthfulness of this explanation (the

boy even peels aside a bandage to show them the angry sore), and fully consistent with

his theoretical distrust ofprima facie evidence, Joll later subjects the two barbarians to

further “questioning,” during which he brutally kills the older man and repeatedly cuts

into the flesh of the young boy with a very small knife, in order, by means of the

traumatic infliction of pain, to extract from him the very “truth” Joll was hoping to hear

(indeed, was sent to the frontierto find) in the first place: that the barbarians were

secretly “arming themselves” and preparing “to join in a great war on the Empire” (10-

11). In the character of Joll, Coetzee makes even more graphic and explicit the

conceptual correlation between epistemological violence and Cartesian pictures of

mental privacy that Rilke, Bacon and Kundera had already suggested.

227

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It would be easy, but only partially correct, to see the liberal-minded Magistrate

as structurally opposed to the torturer Joll in the moral economy of Coetzee’s novel. In

fact, however, the Magistrate’s momentary sympathy for loll (“how natural a mistake to

believe that you can bum or tear or hack your way Into the secret body of the other!”)

hints at the extent to which these two antagonists share the very same concept of inind.

And it is from this common Cartesian ground that spring the Magistrate’s own

“inexplicable attentions” directed towards the barbarian woman, which treat her as an

enigma to be deciphered, a mystery to be comprehended, or better yet, an obdurate shell

to be penetrated in order to access the psychological kernel within. And consistent with

our discussion so far, it is the way he pictures her face that betrays the Magistrate’s

essentially Cartesian understanding of the nature of mind. Having explored how the

treatment of the human countenance in Rilke and Bacon dramatizes the terrible ethical

consequences of a theoretical commitment to mental privacy, we already have a good

grasp of the meaning of the face in Coetzee. Indeed, everything thatI have said about

Rilke and Bacon applies quite directly to the role of the face in Coetzee’s novel, so

much so that I would be surprised if Coetzee did not have Rilke and Bacon in mind

when he wrote those passages that describe the Magistrate’s obsession with the face of

the tortured woman (in fact, I suspect that Coetzee may even have known Kundera’s

L ’Arc essay on the Moraes triptych: the crucial role of state interrogation in Kundera’s

essay, considering Coetzee’s characterization of Joll, is striking).

At one point, deeply frustrated by his feeling that the barbarian woman is

“beyond comprehension,” the Magistrate pictures the features of her face altering in

228

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. terms that bring to mind Bacon’s canvases: “I have a vision of her closed eyes and

closed face filming over with skin,” he thinks. “Blank, like a fist beneath a black wig,

the face grows out of the throat and out of the blank body beneath it, without aperture,

without entry” (42). And reminiscent of Bacon’s predatory approach to painting (“as an

artist, you have to, in a sense, set atrap by which you hope to trap this living fact alive”

[EFB 57]), the Magistrate later complains that “with this woman it is as if there is no

interior, only a surface across which I hunt back and forth seeking entry. Is this how

her torturers felt hunting their secret, whatever they thought it was?” (43). And as if all

this were not disturbing enough, he goes on to identify his erotic “attentions” to the

woman’s body with the interrogation methods used by Joll: “The girl lies in my bed,”

he thinks, “but there is no good reason why it should be a bed. I behave in some ways

like a lover - 1 undress her, I bathe her, I stroke her, I sleep beside her - but I might

equally well tie her to a chair and beat her, it would be no less intimate” (43).

This, I suggest, is Coetzee’s point. There could be nothingless intimate than

thinking of another person’s mind as a hidden, private thing, encased in a shell,

“opaque” and “impermeable,” or like “an urn or a ball,” as the Magistrate variously

describes the woman’s face and mind. On the very next page, he recoils from the

association he had just posited between himself and Joll, but we readers know it is too

late to take the comparison back: “There is nothing to link me with torturers,” he

exclaims to himself. “How can I believe that a bed is anything but a bed, a woman’s

body anything but a site of joy? I must assert my distance from Colonel Joll! I will not

suffer for his crimes!” (44). Of course, it would be absurd (and cynical in the extreme)

229

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. to claim that the Magistrate and loll are really not at all different. Nonetheless, Coetzee

has made clear that the concept of mind that informs the Magistrate’s treatment of the

woman is isomorphic with that which the Empire uses to rationalize the brutal

interrogation practices exercised in the name of state power: the very practices, of

course, that left the woman maimed in the first place and her father dead. It is this that

the Magistrate had suddenly realized, and his later protestations to the contrary,

however forceful, come too late.

And, in any case, when the Magistrate insists that a “bed is... but a bed” and a

“woman’s body... but a site of joy,” the “joy” that he speaks of, which is linked to his

nightly ritual of cleansing and oiling the woman’s body, merely confirms our general

point about the Magistrate’s Cartesian understanding of mind. Here is one typical

description of the ritual: “[I] begin to wash her feet.... I lose myself in the rhythm of

what I am doing. I lose awareness of the girl herself. There is a space of time which is

blank to me: perhaps I am not even present... I push the basin aside and dry the foot...

My eyes close. It becomes an intense pleasure to keep them closed, to savour the

blissful giddiness” (28). The “bliss” the Magistrate experiences here is brought about

by the rhythmic regularity of his hand movements upon the woman’s body, but oddly

enough the pleasure is not at all sensual but rather characterized by a withdrawal from

sense-perception and the experience of time-consciousness. “[Ojften in the very act of

caressing her,” as he says in another description of the ritual, “I am overcome with sleep

as if poleaxed... These dreamless spells are like death to me, or enchantment, blank,

outside time” (31).

230

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Those familiar with that locus classicus of Romantic inwardness, the fifth

promenade of Rousseau’sReveries of a Solitary Walker, will recognize in these

characterizations of the Magistrate’s state of mind a parodic allusion to the “sentiment

of existence” that Rousseau felt while listening to the rhythmic “ebb and flow” of the

waters of Lake Bienne: “I liked then to go and sit... in some secluded spot by the edge

of the lake,” we read in the fifth promenade:

there the noise of the waves and the movement of the water, taking hold of my senses and driving all other agitation from my soul, would plunge it into a delicious reverie in which night often stole upon me unawares. The ebb and flow of the water, its continuous yet undulating noise, kept lapping against my ears and my eyes, taking the place of all the inward movements which my reverie had calmed within me. (86-7)

And as with the Magistrate’s nightly ministrations, these experiences of Rousseau by

the lake provide his soul access to spots, as it were, outside of time: “where time is

nothing to it,” as Rousseau explains, “where the present runs on indefinitely but this

duration goes unnoticed, with no sign of the passing of time... Such is the state which I

often experienced on the Island of Saint-Pierre in my solitary reveries” (88-9)

Like these solitary reveries enjoyed by Rousseau, the rhythmic “pleasure” the

Magistrate enjoys while caressing the woman’s body is solitarya pleasure. Indeed, the

perverse solipsistic experience of the nightly ritual is but the first-person correlate to his

third-person perception of the woman’s face as an “urn or a ball”: both picture mind

itself (whether his or hers) as a monad that is private and closed off from the world.

That is to say, these two textual moments, both the rhythmic bliss and the woman’s

filmed-over face, are vivid expressions of two sides of a single conceptual logic. The

231

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. problem that Coetzee draws our attention to in this novel is the Cartesian picture of

mind, shared by both Joll and the Magistrate, that underlies the entire spectrum of

unethical actions that constitute its narrative arc.Waiting for the Barbarians has often

been referred to as an allegory about the politics of imperialism, and certainly it is that;

but I would urge that we consider it also, and perhaps even primarily, as an allegory

about the philosophy of mind. Coetzee’s novel, I suggest, sees imperialism as an

extension into the field of politics of a mistaken view of intersubjectivity and other

minds that operates first at the level of interpersonal relations between two individuals.

As with Rilke and Bacon before him, though perhaps even more directly and self

consciously, Coetzee’s imaginative representations of the barbarian woman’s face

dramatize, and thereby warn us against, the moral and epistemological pitfalls of our

Cartesian modernity. Beware the seductive picture of mental privacy that is our

common philosophical inheritance, Coetzee’s novel suggests, lest we lose ourselves in

solipsistic fantasies, whether the domestic ones of someone like the Magistrate, or

imperialistic ones such as those of Colonel Joll. Either way, we risk overlooking the

truth about those around us, perhaps simply (and somewhat surprisingly) because we

insist on looking too hard. The Magistrate describes well the sort of self-blinding gaze

(this “looking too hard”) that he suffers from when, in the following passage, he thinks

about how he hopes the barbarian woman regards his obsessive attention: “I have

hitherto liked to think that she cannot fail to see me as a man in the grip of a passion,

however perverted and obscure that passion may be, that in the bated silences which

make up so much of our intercourse she cannot but feel my gaze pressing in upon her

232

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with the weight of a body” (56). It is the very intensity of his gaze—intent on

fathoming her psychological depths—that obscures from sight whatever might simply

be there to be seen on the surface in the first place.

Near the end of the novel, the Magistrate thinks sadly to himself: “There has

been something staring me in the face, and still I do not see it” (155). But is it not the

woman’s face itself that has been staring him in the face, as in the dream sequences that

he obsessively revisits night after night? Is it not the face itself that he does notseel

Which he cannot see because he refuses to take her at face value, since (like Joll) he

believes that the truth about persons must be hidden, buried, and mysterious? At

various moments throughout the course of the novel, the Magistrate—relentlessly self-

reflective as he is—insightfully glimpses (however dimly and briefly) his own

complicity in this fundamentally projective and fantasmatic interpersonal dynamic, in

which he “constructs” the barbarian womanas an enigma to be deciphered. It is not

she who is “without aperture, without entry,” but he who insists on seeing and treating

her that way (42). It is not she who has “no interior, only a surface across which I hunt

back and forth seeking entry,” but he who approaches her in that way, and then, in

exasperation, expresses sympathy for those, like Joll, who make the “natural” mistake

of believing that one can “hack” one’s way into the “secret body of the other” (43). In

one of his passing moments of lucidity, the Magistrate revealingiy characterizes his

treatment of the woman this way: “So I continue to swoop and circle around the

irreducible figure of the girl, casting one net of meaning after another over her” (81).

The Magistrate, in other words, senses what he is doing to the “girl”: that he is treating

233

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. her as a theoretical object suited to bear his meanings rather than express her own

(though it is essential to this psychological dynamic that he believe heis trying to

understand her on her own terms: that is. he imagines her to have a deeply hidden self

which he then must forcefully struggle to disclose). Perhaps all this can make sense of

the following recurring dream, in which what one might call the semantic fluidity or

plasticity of the girl-figure in the Magistrate’s psychic economy becomes terrifyingly

evident: “There are other dreams in which the figure that I callthe girl changes shape,

sex, size. In one dream there are two shapes that arouse horror in me: massive and

blank, they grow and grow till they fill all the space in which I sleep. I wake up choked,

shouting, my throat full” (87; italics in original). This nightmarish dream makes all too

clear that “the girl” that inhabits the Magistrate’s thoughts is not the barbarian woman

at all, but something like pure unformed matter, indeterminate, suited to take whatever

conceptual determination best meets the Magistrate’s theoretical compulsion to imagine

mind as such as radically private. It is this “massive and blank” formlessness—which

brings to mind the “bare flayed head” of the faceless woman in Rilke’sNotebooks —and

not some true hidden essence, that lies “behind” or “beneath” the filmed-over face the

Magistrate pictures to himself when he thinks of the woman.

Derek Abridge, in his recent monograph on Coetzee, sees the significance of the

girl and the dreams quite differently.108 Opposing a critical tendency to readWaiting

for the Barbarians in an allegorical fashion, he argues that the difficulties that the

dreams pose for the Magistrate’s (and the reader’s) drive to interpret should be seen as

108 Derek Attridge, J.M. Coetzee and the Ethics o f Reading: Literature in the Event (Chicago: Univ. of Chicago Press, 2004).

234

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. an ethical lesson about the virtues of semantic uncertainty and open-endedness.

According to Abridge, it is in “the girl’s” very resistance to signification that her true

significance lies. That is, the Magistrate’s “failure to interpret” gives us access to a

literary and intersubjective experience that is “important” and “emotionally powerful”

precisely because it is, and must be, a failure:

Take the dreams: there are several in the novel, including a series of related dreams set in the town square, usually seen as snow- covered and always involving a figure that the Magistrate refers to as “the girl.” .... The reader is, of course, invited to interpret them, and to relate them to aspects of the Magistrate’s waking life (such as his frequent vision of the barbarian girl’s face as blank), but part of their haunting power derives from their refusal to succumb completely to the interpretive drive.... The point I wish to make is that allegorical reading of the traditional kind has no place for this uncertainty and open-endedness, this sense that the failure to interpret can be as important, and quite as emotionally powerful, as success would be. (47-48)

I have no complaint with Attridge’s dissatisfaction with allegorical readings of the

“traditional kind,” which, as he describes them further, attempt to definitively link

elements within the novel to “parallels outside the world of the book” (48). But as I

hope is clearly evident from my reading of the novel so far, I believe that his

explanation of the significance of “the girl” misses Coetzee’s point. Indeed, Attridge’s

reading badly distorts a matter of fundamental importance to the novel, for by

championing the girl’s resistance to interpretation as an ethical virtue, he is actually

identifying with the Magistrate’s Cartesian prejudices, the very ones that originally

transformed her into an incomprehensible entity that so obdurately resists interpretation

in the first place. If she were actually as enigmatic or opaque as the Magistrate finds

235

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. her to be, then Attridge would have a point: but the real point of Coetzee’s

characterization of the barbarian woman is that the Magistrate, by casting his “nets of

meaning” upon her, hasmade her otherness into a metaphysical problem. That “the

girl” seems so resistant to the Magistrate’s interpretive gaze is not evidence of the

irreducibility and value of alterity, but the result of a solipsistic constraal of the problem

of other minds.

What the Magistrate blinds himself to in his rush to penetrate to the barbarian

girl’s hidden psychic essence is necessarily unclear, since this novel about solipsism is

appropriately and strategically narrated in the first-person. Like him, we have no

privileged access to the minds and thoughts of others. Even still, Coetzee provides us

with enough clues to suggest rather clearly how badly mismatched are the Magistrate’s

thoughts and the intersubjective reality they intend to interpret. For example, consider

the interaction in the following passage, in which it becomes clear that the Magistrate

would prefer the barbarian woman to remain an interpretive enigma rather than assert

herself as a real flesh-and-blood presence with thoughts, emotions and desires of her

own:

“Wouldn’t you like to do something else?” she asks. Her foot rests in my lap. I am abstracted, lost in the rhythm of rubbing and kneading the swollen ankle. Her questions takes me by surprise. It is the first time she has spoken so pointedly. I shrug it off, smile, try to slip back into my trance, not far from sleep and reluctant to be diverted. The foot stirs in my grip, comes alive, pokes gently into my groin. I open my eyes to the naked golden body on the bed. She lies with her head cradled in her arms, watching me in the indirect way I am by now used to, showing off her firm breasts and her sleek belly, brimming with young animal health. Her

236

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. toes continue to probe; but in this slack old gentleman kneeling before her in this plum dressing-gown they find no response. “Another time,” I say, my tongue curling stupidly around the words. As far as I know this is a lie, but I utter it: “Another time, perhaps.” Then I lift her leg aside and stretch out beside her. “Old men have no virtue to protect, so what can I say?” It is a lame joke, poorly expressed, and she does not understand it. She slips open my gown and begins to fondle me. After a while I push her hand away. (55)

The Magistrate’s refusal, which even he regards as rather lame, understandably pains

the woman, who heatedly reproaches him for deflecting her advances when she knows

that he regularly visits “other girls”: ‘“Do you also treat them like this?’ she whispers,

and starts to sob” (55).

Sensing the turn that this quarrel marks in their relationship, though not

understanding it, the Magistrate sees to it that this “is the last night [they] sleep in the

same bed” (55). It is important to appreciate that the Magistrate does not clearly grasp

his own obscure erotic motivations: he, who has been an easy and promiscuous lover

his entire adult life, has no explanation for why he does not desire to have sex with this

particular woman, though she herself expresses a desire to do so. After having returned

one night to the bed he shares with the barbarian woman after a visit to a favorite

prostitute (the one he calls the “Bird”), the Magistrate ponders this very question,

asking after the perplexing nature of his desire for the barbarian woman:

The girl I have just left, the girl she may perhaps (I suddenly realize) smell on me, is very pretty, there is no question about that: the acuteness of my pleasure in her is sharpened by the elegance of her tiny body, its manners, its movements. But of this one there is nothing I can say with certainty. There is no link I can define between her womanhood and my desire. I cannot even say for sure that I desire her. All this erotic behavior of

237

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mine is indirect: I prowl about her, touching her face, caressing her body, without entering her or finding the urge to do so. 1 have just come from the bed of a woman for whom, in the year I have known her, I have not for a moment had to interrogate my desire: to desire her has meant to enfold her and enter her, to pierce her surface and stir the quiet of her interior into an ecstatic storm; then to retreat, to subside, to wait for desire to reconstitute itself. But with this woman it is as if there is no interior, only a surface across which I hunt back and forth seeking entry. Is his how her torturers felt hunting their secret, whatever they thought it was? (43)

The Magistrate’s “interrogation” of his own desire is revealing but inconclusive: about

the barbarian woman “there is nothing [he] can say with certainty.” It is not that he

finds her w/tdesirable. Rather, it appears that his desire takes a different direction: not

aiming to enter her literal body but rather to penetrate to something mysterious (like the

“secret” her torturers sought) that he assumes is hidden deep within her mind.

Strikingly, however, though his desire is not literally carnal, he imagines his goal (her

“secret”) in corporeal terms: the psychological distinction between inner and outer is

rendered literally palpable (later he will also imagine even his own “gaze” as “pressing

in upon her with the weight of a body”). But so mysterious is this secret of hers—this

secret that she has become for him—that he can perceive her only as a surface (“as if

there is no interior”): obdurate, enigmatic, unyielding. It is unsurprising then, that it is

immediately following this very passage that he experiences his feeling of “dry pity” for

torturers like Joll who attempt to “tear” their way into the “secret body of the other”

(43). For it is at moments like this, when he imagines her as holding a secret that must

be disclosed, that he reveals how deeply he shares Joll’s Cartesian theory of mind.

238

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. So Intent Is he on gaining “entry” Into her “secret body” that he rashly overlooks

the way she expresses her mind, quite dearly with her actual body, whether with her

actions, words, or tears. It Is as though he were following M l’s method for torture, by

methodically rejecting allprima facie evidence, in order to get at a latent and private

inner truth (and Is it not clear that the Magistrate’s method of “interrogation” thereby

inflict its own kind of pain as well?). Near the very end of the novel, after he has been

sleeping with Mai, the cook at the town’s inn, she remarks to him that he always seems

distant in bed, just as the barbarian woman used to say he often was with her:

“I don’t make you happy,” she says. “I know you don’t enjoy it with me. You are always somewhere else.” I wait for her next words. “She told me the same thing. She said you were somewhere else. She could not understand you. She did not know what you wanted from her.” “I didn’t know you and she were intimate.” “I was often here, downstairs. We talked to each other about what was on our minds. Sometimes she would cry and cry and cry. You made her very unhappy. Did you know that?” She is opening a door through which a wind of utter desolation blows on me. “You don’t understand,” I say huskily. She shrugs. I go on: “There is a whole side to the story you don’t know, that she could not have told you because she did not know it herself. Which I don’t want to talk about now.” (152)

The testiness of his final response indicates how deeply Mai’s remarks have cut him,

that indeed (as melodramatic as mit a y sound) it is as though a door had been opened in

his monadic ego through which the painful truth about the outer world were rushing in.

But has Coetzee not made it clear that if the Magistrate did not appreciate the pain he

was causing the barbarian woman, it was simply because he refused to acknowledge it?

239

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Far from being secreted away, her state of mind was readily accessible by the most

mundane and straightforward of means (as Mai observed, “Wetalked to each other

about what was on our minds”!). Between the Magistrate and the woman, however, it

was not speech but “bated silences” that constituted their form of “intercourse,” lest talk

interfere with his attempts to enter her “secret body” with his “gaze” But could

anything in Mai’s remarks truly have surprised him? The desolation the Magistrate

feels as he listens to her might just as well derive from the shame of self-recognition as

it could from the shock of some radically new knowledge. It is probably a bit of both.

So though I do not dispute the truism that interpretation must necessarily remain

uncertain and open-ended, I cannot agree with Abridge’s argument that the Magistrate’s

“failure to interpret” the dreams or “the girl” constitutes some sort of positive ethical

lesson of the novel: that “the failure to interpret can be as important, and quite as

emotionally powerful, as success would be” (48). There are clearly limits to what we

can know about one another, which at times can be painful to accept, but these limits

are not those that Coetzee makes the Magistrate confront. I hope I have demonstrated

that Coetzee’s point is not that we, like the Magistrate, must come to terms with the

irreducible opacity of the barbarian woman. On the contrary, if the Magistrate had not

insisted on imagining the girlas an enigma, he might have been able to see, and

acknowledge, their common humanity (by which I mean nothing metaphysically

elevated, but only such everyday things as thoughts, emotions, and desires). In recent

decades, the related concepts of “otherness” and “difference” have been elevated to

theoretical and political fetishes of sorts, but we do an injustice to Coetzee’s text by

240

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. imagining that it reflects or participates in this intellectual, trend. Far from insisting on

the woman’s radical alterity,Waiting for the Barbarians strongly suggests that we are

not nearly as different from one another as we often like to imagine. By identifying

with the Magistrate’s frustrated hermeneutic point of view, Attridge shows how difficult

a lesson this can be to leam.

• Levinas: Phenomenology and the Other

Levinas, whose philosophical writings have played a major role in focusing

critical attention onto the concept of alterity in recent decades, might very well have

sympathized with Attridge’s interpretation of Coetzee’s novel: in fact, it is probably the

mediated influence of Levinas’s own ideas about otherness (transmitted by way of his

influence upon Derrida) that informed Attridge’s interpretation of Coetzee in the first

place. Having considered how Coetzee’s Magistrate solipsistically transforms the

barbarian woman into a psychic enigma that mirrors his own Cartesian picture of

mental privacy, it is time finally to return to Levinas’ ethics of the face. Levinas would

regard the Magistrate’s mistreatment of the tortured woman as a typical instance of

western thought’s stubborn inability to think of another person without ensnaring him or

her in “nets of meaning” woven by the intentional consciousness of the sense-bestowing

self.

241

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The phenomenological investigations of Levinas5 teacher, Husserl, had made

clear that mental life as a whole is characterized by object-directedness or intentionality,

a fundamental and far-reaching claim that Levinas explains as follows:

The famous proposition that “all consciousness is consciousness of something,” or that intentionality essentially characterizes consciousness, sums up the Husserlian theory of mental life: every perception is perception of a perceived, every judgment is judgment of a state of affairs judged, every desire is desire for a desired. This is not a correlation of words, butdescription a of phenomena. At all levels of mental life, whether at the stage of sensation or of mathematical thought, thought is aim and intention. (DEH 58; italics in original)

Phenomenology is intended to be radically presuppositionless, producing “descriptions”

(as Levinas emphasizes) of how things appear to us in consciousness rather than

formulating theoretical explanations based upon metaphysical or scientific claims about

the nature of perception or the physical structure of the material world. As Dan Zahavi

explains this aspect of phenomenological method:

Husserl wants to describe our experiences as they are given from a first-person perspective, and it is no part of my experience of, say, a withering oak tree, that something is occurring in my brain. Thus, already early on Husserl stresses the (metaphysical) presuppositionlessness of phenomenology. Phenomenology is supposed to be neither more nor less than a faithful description of that which appears (be it subjective acts or worldly objects), and should, as a consequence, avoid metaphysical and scientific postulates or speculations. (HP 13-14; italics in original)

Phenomenological description thus begins with our conscious experiencesas they are

experienced:nothing that cannot be validated in one’s own experience is to be accepted.

This includes even the ordinary (but, in fact, “metaphysical”) assumption that a real

world, external to the mind, exists in itself. All theories, however seemingly innocent,

242

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. are to be validated by experience; experiencesnot are to be interpreted with the aid of

theories. As Dermot Moran puts it, “Phenomenology is a return to ‘phenomena’.

Husserl understandsphenomenon as ‘what appears as such’; in other words, everything

that appears, including everything meant or thought, in themanner of its appearing, in

the ‘how’ (Wie) of its manifestation” (IP 127). Or as Heidegger would sum up this

aspect of Husserl’s teachings in a well-known passage ofBeing and Time : “Thus

‘phenomenology’ meansapophainesthai ta phainomena - to let that which shows itself

be seen from itself in the very way in which it shows itself from itself’ (BT §7).

Husserl encapsulated his methodological stress on the presuppositionlessness of

phenomenological description in his famous “principle of principles”Ideas in I §24:

Enough now of absurd theories. No conceivable theory can make us err with respect to the principle of all principles: that every originary presentive intuition is a legitimizing source of cognition, that everything originarily (so to speak in its “personal” actuality) offered to us in “intuition ” is to be accepted simply as what it is presented as being, but also only within the limits in which it is presented there. (IP 127; italics in original)

In order to purify oneself of all “absurd theories” that might prejudice or contaminate

one’s descriptions of mental life, Husserl recommended a set of procedures he called

“reductions” (from the Latinreducere, “to lead back”) that were designed to lead us

back, as it were, from received theory to naked experience. Moran explains the

motivation for the reductions this way:

Husserl had a number of different theoretical reasons for introducing the notion of reduction. First it allowed him to detach from all forms of conventional opinion, including our commonsense psychology, our accrued scientific consensus on issues, and all philosophical and metaphysical theorising

243

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. regarding the nature of the intentional. We must put aside our beliefs about our beliefs, as it were. Secondly, it allowed him to return to and isolate the central structures of subjectivity. By putting aside psychological, cultural, religious, and scientific assumptions, and by getting behind or to one side of the meaning- positing or thetic acts normally dominant in conscious acts, new features of those acts come to the fore. (IP 146)

What Husserl found left in his consciousness, when by means of his various reductions

he had purified himself of all metaphysical, scientific and religious presuppositions, was

that consciousness was always consciousnessof: that is, that consciousness was

characterized by object-directedness or what (barkening back to a Scholastic term of art

revived by Brentano) he also called “intentionality.” As Zahavi explains the object-

directedness of the intentional relation: “Regardless of whether we are talking of a

perception, thought, judgment, fantasy, doubt, expectation, or recollection, all of these

diverse forms of consciousness are characterized by intending objects and cannot be

analyzed properly without a look at their objective correlate, that is, the perceived,

doubted, expected object” (HP 14). Husserl discovered that once one grasps

consciousness as thoroughly intentional, the characteristic modem rift between subject

and object is healed, for every intentional correlation contains as an essential part of it

both intending act and intended object. Though these two components (which he would

later term “noesis” and “noema”) are distinguishable in theory, in actuality they

constitute two inseparable poles ofone noetic-noematic correlation. Thus, my (noetic)

perceiving of a tree is bound inextricably to the (noematic) tree that I perceive. Though

act and object can be artificially teased apart for theoretical consideration, there is no

actual rift between the subjective act and the object perceived. As Levinas put it,

244

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. phenomenology “leads as outside the subject-object categories and topples the

sovereignty of representation. Subject and object are only the poles of this intentional

life” (DEH 119).

What is important to keep in view, however, is that this “intentional life” in

which the metaphysical distinction between subject and object is transcended is one that

is strictly immanent to the consciousness of the intending self. As I noted earlier, one of

the various “presuppositions” that Husserl’s reductions were designed to bracket out

was the “metaphysical” belief that an objective world exists in itself, regardless of our

consciousness of it. After the phenomenological reduction, the “reality” of the

objective world remains, of course, but onlyas an intentional objectimmanent to the

phenomenologically purified consciousness. The “objectivity” of the world survives

the reduction, that is, but the reduction helps us to see its objectivity as nothing other

than a meaning the world hasfor me. The same is obviously true for any phenomena

that has the “appearance” of objectivity: after all, the very idea of appearance is

inherently indexed to a subjective consciousness. A phenomenologically clarified

understanding of “objectivity,” therefore, shows it to be nothing but the manner in

which certain phenomenaimmanentiy appear to a subjective consciousness as

transcendent to it, and it is in this way that the modem rupture between subject and

object is “healed” by the phenomenological discovery of intentionality.

Because all phenomena are understood as immanent, and essentially indexed, to

an intending consciousness, Husserl also often said that whatever meanings they have

are therefore “constituted” by the intending self. Since they do not have their meanings

245

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. in themselves, where else could their meaning come from? Meaning, according to

Husserl, can only be meaning/orus, and so it is simply a different way of expressing

this same point when he claims that meaning is given or constitutedby us. As Moran

explains the important phenomenological notion of “constitution”:

For Husserl, “constitution” expresses the manner in which objects of consciousness come to have the kinds of “sense and being” that they do, the manner in which subjectivity carries out its function of giving sense. Husserl’s notion of constitution should perhaps be thought as a kind of setting out or “positing” (,Setzung), as a giving of sense, “sense-bestowing”(Sinngebung). (IP 164-5)

Or as Levinas put it, the “relation of intentionality is nothing like the relations between

real objects. It is essentially the act of bestowing a meaning (the Sinngebung)” (DEH

59; italics in original).

I should pause to note that the reading of Husserl’s theories of intentionality and

constitution that I have been developing is a contentious one, tending to portray his

understanding of constitution as more idealist in nature than many of his supporters

would accept. For instance, Zahavi (citing Heidegger in support), argues against

idealist interpretations of constitution and instead suggests that we should rather regard

constitution as a way of letting objects show themselves as what they are:

Constitution must be understood as a process that allows for manifestation and signification, that is, it must be understood as a process that permits that which is constituted to appear, unfold, articulate, and show itself as what it is. As Heidegger was to observe: “‘Constituting’ does not mean producing in the sense of making and fabricating; it meansletting the entity be seen in its objectivity” (HP 73; italics in original)

246

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. However, there is room for interpretive disagreement here. Even though Moran himself

does not support an idealist reading of the notion of constitution, he nonetheless points

out that Husserl “does actually speak of transcendental consciousness giving both

meaning andbeing to the world, but ‘being’ here means the manner in which beings

appear to consciousness, being-for-us as opposed to being-in-itself (terms Husserl

himself employs)” (IP 165). Summarizing his own positive and supposedly non-idealist

interpretation, Moran writes:

Husserl’s stress on constitution does not rule out the recognition of the facticity of the world, and the manner in which contents appear in consciousness over which it has no control. Constitution includes a kind of passive construction of all the meanings found in consciousness. Rather the whole object as such is experience asgiven from the world. (IP 166)

It is unclear to me how far this actually differs from an idealist reading of Husserl: after

all, even an idealist understanding of constitution would acknowledge that external

objects are experienced “as given from the world,” as Moran puts it. “Facticity” might

simply be a meaning objective reality has for us, rather than in itself. But any attempt at

reaching a definitive interpretation of the Husserlian concept of constitution is far

beyond the scope of this study. As Zahavi says, Husserl himself “never gave a clear-cut

answer to the question of whether constitution is to be understood as a creation or a

restoration of reality,” and thus “one of the recurrent problems in Husserl research has

been the question of how exactly to understand Husserl’s notion of constitution,

particularly in its implications for the discussion between realism and idealism” (HP

72). If I have chosen to emphasize the idealist way of reading Husserl’s theory of

247

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mental life, I have done so not only because I tend to find it persuasive, but more

importantly, because it is clearly the view taken by Levinas, and it is this interpretation

of Husserl that fundamentally determined the development and eventual shape of his

own later phenomenology of the face.

According to Levinas’ reading of Husserl, that is, intentional consciousness

cannot help but understand the Other according to terms originating in what Levinas

calls the Same (that is, the constituting “primordial sphere” of the monadic

1 fK) transcendental ego). Thus, when I perceive another person, I constitute him or her in

my mind as a kind ofalter ego: in other words, a differentme (somewhat like

Augustine’s analogical inference to the existence of other minds). Intentionality, as

Levinas put it, is “essentially the act of bestowing a meaning (theSinngebung).” That

which is truly Other than the Same, by definition absolutely—or as Levinas liked to say,

“infinitely”—exterior to it, by being made the intentional object of the consciousness of

the self, is reduced to a mere structural moment in the closed conceptual totality that

characterizes the essentially monadic consciousness of the transcendental ego: hence the

antinomic terms that make up the title of one of Levinas’ major texts,Totality and

Infinity. In this difficult work, Levinas would defend the ethical claims of the infinitely

Other against the encroachments of all forms of theoretical (that is,intentional)

totalization.

109 On Husserl’s use of the Leibnizian term “monad,” Dermot Moran writes: ‘The ‘monad’ is Husserl’s name for the whole concrete conscious life of an ego taken as the full set of all its intentional experiences, both actual and possible (CM § 33, 68; Hua 1 102). It is the complete draft of a life as it were. Husserl speaks of ‘monadisation’ of the transcendental ego and of the self as a ‘monad with windows’” (Moran 174).

248

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Obviously, Levinas believes that it is somehow possible to come into actual

contact with radical alterity, otherwise the complaints he levels against the

phenomenological tradition would remain purely formal and empty. A form of

“alterity,” however conceivable, that had no role in our lived experience—that was

simply an abstract idea—could exercise no obligatory force upon us and should

therefore have no bearing on our ethical thought or conduct. But Levinas insists that we

can indeed experience (and not simply think about or imagine) absolute alterity, and he

argues that the “infinitude” defining the otherness of the other can be directly sensed in

the appearance of his or her face: “the gleam of exteriority or of transcendence in the

face of the Other” (T I24). Indeed, it would be more accurate to say that the radical

alterity Levinas speaks of could only be “experienced” and not ever faithfully “thought”

in the first place, since any “idea” of the other would already be an objectified ideality,

a concept (and therefore absorbed into the totality of the Same). But if alterity can

somehow be directly experienced, how do we “see” this “gleam” of the face, since

Levinas appears to accept the idealist findings of phenomenology, from which it

follows that “in visual or tactile sensation the identity of the I envelops the alterity of

the object” (TI 194)? In other words, like all forms of consciousness and thought,

perception is intentional in nature, and works by synthesizing together various

perspectivally given adumbrations of perceptual objects into cognitive unities (“The

identity of a unity across multiplicity represents the fundamental event of all thought.

For Husserl, to think is to identify,” notes Levinas [DEH 59]). Because the face of the

other, when approached in an ethical manner, cannot (according to Levinas’ definition)

249

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. be intended like any ordinary object of cognition, whatever epiphanic “vision” we

might have of the other’s face must therefore be highly peculiar.

In his preface to Totality and Infinity, Levinas summarizes as follows the

paradoxical phenomenology of “vision” (which he tellingly associates with

eschatoiogy) that informs his understanding of the face-to-face relation:

The first “vision” of eschatoiogy (hereby distinguished from the revealed opinions of positive religions) reveals the very possibility of eschatoiogy, that is, the breach of the totality, the possibility of a signification without a context. The experience of morality does not proceed from this vision—itconsummates this vision; ethics is an optics. But it is a “vision” without image, bereft of the synoptic and totalizing objectifying virtues of vision... (TI 23, italics in original)

“Ethics is an optics,” we are told, but the “vision” of the other’s face that opens, indeed

“consummates,” the ethical relation between two persons is perceptually empty:

“without image.” That is to say, the face of the other, when seen aright, is not seen at

all. The face-to-face encounter has nothing to do with perception (facial or otherwise),

and indeed Levinas repeatedly associates sensory perception with epistemological

violation: vision is “totalizing” and “objectifying,” as he puts it here. In other words,

Levinas’ ethics of the face is essentially iconoclastic: as with the face of God in the

Hebrew Bible, one respects the otherness of the Other by refusing to draw it into the

field of representation. But Levinas’ phenomenological understanding of the reach of

representation is wide indeed:all “ mental life participates in representation,” as he

wrote in one of his essays on Husserl (DEH 59; emphasis added). Therefore, according

to his own iconoclastic logic, the unrepresented and ethicallyunrepresentable otherness

250

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the Other must lie outside the bounds of cognition (“mental life”) as a whole. If

Levinas nonetheless insists that ethical contact with the Other is possible, it can only be

so beyond perception, intentionality, and thus thought itself. No wonder he thought of

the epiphany of the face as a kind of eschatological vision: to take Levinas’ notion of

the face seriously is to entertain the possibility of a revelation beyond the scope of

philosophy, reason and even thought as such.

And yet he did not want to see himself as having abandoned philosophy for

religion: as is well known, he repeatedly insisted on a strict difference in his writings

between his philosophical reflections and his Talmudic interpretations. Levinas argued

that he had discovered philosophicala explanation for how we can sense the alterity of

the other person in Husserl’s various writings on time-consciousness, particularly in the

concept of the primal impression Urimpression( ) that plays such an important role in

Husserl’s analyses of the experience of time. That is, Levinas believed he had

discovered within Husserl’s own philosophical system a description of an experience

that exploded the egological bias of phenomenology. In finding such rich conceptual

possibilities in Husserl’s writings on time-consciousness, Levinas was far from alone:

Heidegger before him, and Derrida after, are just two of the many phenomenologically

oriented thinkers who based their own original philosophies upon what they believed

were untapped conceptual possibilities latent in Husserl’s probing reflections on time.

The distinguished Husserl scholar Rudolf Bemet notes the irony of this philosophical

legacy:

251

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. It is almost as if the Husserlian descriptions of the experience of time contained within themselves the seeds of a surpassing of the philosophical framework in which Husserl had inserted them. We are then confronted with the paradox whereby an analysis of time that was to have provided a foundation for a phenomenology of an egological transcendental consciousness constitutive of objects by justifying their epistemological validity also retains a large part of its value in anontological phenomenology ofDasein or in anethical phenomenology of the other person who appears in the form of the ‘face’ or the ‘appeal’. (CCL 82)li0

But what philosophical and ethical possibilities did Levinas in particular discern in

Husserl’s writings on time-consciousness, especially in his conception of the primal

impression? In order to understand and evaluate Levinas’ challenging reading of

Husserl’s already dense reflections on time-consciousness, however,I think it would be

helpful to first sketch out the basic outlines of Husserl’s own views.

Like many of his fellow modems (such as William James and Henri Bergson),

Husserl considered constitutive consciousness to be temporal in nature: consciousness,

as he liked to say, is a constantly streaming flux(Fluss). According to Husserl, every

phase of consciousness has a tripartite temporal structure: (1) the primal impression (our

consciousness of the now-moment), (2) a retentional moment (the continuing presence

in consciousness of a prior now-moment), and (3) a protentional moment (the present

expectation or anticipation of a now-moment yet to come). Though these three

“moments” can be theoretically distinguished, they are actually inextricably woven

together in each lived phase of consciousness. The constantly streaming present, past

and future are all aspects of a single present, or as Dan Zahavi aptly puts it, Husserl

li0 The Cambridge Companion to Levinas, Eds. Simon Critchley and Robert Bemasconi (Cambridge: Cambridge Univ. Press, 2002). References to this text will be parenthetically cited as CCL.

252

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. insists on “thewidth of presence” (HP 82; my emphasis). That is to say, the temporal

present is not punctual In nature. We can therefore perceive temporal objects, such as

musical melodies, because “consciousness Is not caught in the now,” contrary to the

view' of thinkers like Brentano, who argued that we can indeed immediately “perceive”

each punctual now-moment but can onlyimagine the immediate past and future.

Husserl, arguing against Brentano’s position, held that consciousness is clearly able to

directly perceive—as an essential part of a “thick” present—notes that have faded into

the past as well as anticipate notes that will come in the near future. As Bemet

explains:

It is not for no reason that Husserl speaks of the ‘flux’(Fluss) of consciousness. The rhythm of this flux is articulated by the emergence of a new intentional act succeeding the previous one, which is thus pushed into the past. In most cases, this new act is not without links to the previous act; it was already present in the form of an anticipated future before being effectively realized in the present. And what is true of the act in its temporal duration is equally true of each instant within this duration: consciousness of the present is always intertwined with consciousness of the past and of the future, and this is the very reason why consciousness is a flux and not a succession of separate punctual instants. (CCL 84)

It is within the context of such a tripartite temporal horizon that objects which endure in

time are intended and constituted as cognitive unities. Indeed, absent this temporal

horizon, we would only be able to experience a series of disconnected and punctual

“nows.” Even the simple visual perception of physical objects, such as tables, can only

be accomplished because the synthesizing powers of consciousness, which work by

uniting various perspectival views of an object, are not limited to the disjointed

253

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. perception of punctual “snapshots.” Object-constitution, that is, presupposes that

consciousness can reach through, and synthesize together, individual appearances and

perceptual acts extended in a temporal manifold. Thus, as Zahavi notes, “temporality

must be regarded as the formal condition of possibility for the constitution of any

objects,” indicating one reason for the immense importance Husserl attributed to his

analyses of time-consciousness (HP 80).

What Husserl came to realize, however, is that not only are transcendent objects

(such as melodies) constituted in a temporal fashion, but that ourconsciousness of our

own subjective awareness of these objects (the experience of thehearing of the melody,

for instance) is temporally constituted as well. As Zahavi explains: “Our acts and

experiences are themselves temporal unities which arise, persist, and perish. They are

also constituted in a network of primal impressional, retentional, and protentional

intentions, and are only given, only self-aware within this framework” (HP 85).

Therefore, we are aware not only that a melody we have just heard “ages” or “fades”

with the passing of time, but that the various subjective acts that constituted our

awareness of the melody also age with the temporal streaming that characterizes our

mental life (once the melody has passed, we don’t simply “remember” the melody

itself, but our awareness of the actual experience of aural perception palpably fades too

with the passage of time). As Husserl puts it: “Hence the flow of consciousness

obviously becomes constituted in consciousness as a unity too. The unity of a tone-

duration, for example, becomes constituted in the flow, but the flow itself becomes

constituted in turn as the unity of the consciousness of the tone-duration” (EH 216).

254

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Husserl thus found it necessary not only to analyze the experience of time that underlies

our constitution of objects, but also the experience of time that underlies the self

givenness (or self-constitution) of subjectivity itself.

Yet, Husserl discovered that he could not simply explain the temporal

constitution of subjective acts of awareness in the same way that he had explained the

constitution of transcendent temporal objects, for this would lead to an infinite

explanatory regress. Consciousness, that is, cannot be temporally constituted in the

very same way as transcendent temporal objects (such as melodies) unless something

stands to our constituting-consciousness as our constituting-consciousness stands to the

temporal objects, in which case we would then have to ask what constitutes that which

constitutes our consciousness (and so on and so forth). In order to block the possibility

of this kind of dizzying conceptual regress, Husserl felt compelled to posit an absolute

consciousness that stands outside the constituted stream of time itself; in fact, Husserl

concluded that this absolute consciousness is itself the origin of the temporal stream of

constituting-consciousness in the first place. It is worth reading the following gloss by

Zahavi, who explains this difficult idea with characteristic clarity:

The first and decisive point to make is that our consciousness of that which is given in constituted time... is not itself given in the same kind of time, since this would lead to an infinite regress. If time-constituting consciousness were itself given in constituted time, it would be necessary to posit yet another higher-order time-constituting consciousness, and so forth. It is for this reason that Husserl denies that the time-constituting consciousness, the absolute stream as he also calls it, is simultaneous with that which is temporally constituted. To speak of simultaneity is to posit a common temporal denominator, which is exactly what has to be avoided. The stream is not influenced by temporal change;

255

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. it does not arise or perish in objective time, nor does it endure like a temporal object. Occasionally, Husserl will speak of the stream as if it were atemporal or supratemporal, but this should not be misunderstood. The stream is atemporal in the sense of not beingin time, but it is not atemporal in the sense of lacking any reference to time. On the contrary, the stream is always present, and this standing now(nunc stans) of the stream is itself a kind of temporality. (HP 86-7)

What Levinas found profoundly suggestive in Husserl’s writings on time-consciousness

is that Husserl placed at the origin of this streaming absolute consciousness a primal

impression (Urimpressiori) which, because it is somehow “prior” to the constitution of

inner time as such, is not itself an object of an intentional relation (as we noted earlier,

intentionality presupposes a temporal horizon). Emerging at what Husserl terms the

“originary source point” (Urquellpunkt) of absolute consciousness, the primal

impression thus occupies a somewhat odd and anomalous position in the

phenomenological system: without it, there can be no intentional consciousness, and yet

the Urimpression itself cannot be intended and thus is not a phenomenal part of the

consciousness to which it gives rise. Remarkably, Husserl felt compelled to found

phenomenology itself (conceived of as purely descriptive) upon a conceptual item that

was explicitly non-phenomenal and thus beyond description (it is surely worth asking

whether the very positing of anUrimpression does not violate Husserl’s “principle of

principles,” thus undermining the foundation of his entire theoretical edifice, but that is

a question that will have to be left for another day). Husserl explained that this

Urimpression could be intended only once it had been “pushed” into the retentional past

by the emergence of a new Urimpression, thereby giving rise to the possibility of both

256

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. inner time and self-consciousness; prior to that elapsing, however, the primal

impression is strictly beyond the reach of objectifying intentionality (i.e. consciousness

as a whole).

The role of the Urimpression at the atemporal origin of absolute consciousness,

Levinas argued, showed that “consciousness is delayed in relation to itself’ (DEH 144):

Husserl himself had expressed a similar thought when he wrote that, “We therefore owe

it to retention that consciousness can be made into an object” (EH 220). In Husserl’s

Urimpression (Levinas calls It the “proto-impression”), Levinas claimed to have found,

at the very origin of phenomenological subjectivity, a phenomenon that was irreducible

to the subject-object dichotomy that characterized the objectifying intentionality of

subjective mental life. In the Urimpression, Levinas argued, “perceived and perceiver

are simultaneous,” and thus the “proto-impression alone is pure of all ideality” (DEH

144). If, as Levinas argued, the meaning or sense of all intentional objects are

idealistically bestowed by the consciousness that intends them, then at least in this

proto-impression he had found a “phenomenon” (which he would term the face

[visage]) that transcended (even if only “momentarily”) the violent reach of the

meaning bestowal(Sinngebung) of intentionality. It is worth noting that so much, and

no more, is justified by Husserl’s analysis of time-consciousness: that is, Husserl shows

us that at the origin of absolute consciousness is a phenomenon that lies outside the

bounds of consciousness and intentionality, period. But precisely because this

“phenomenon” lies beyond the reach of intentionality (and therefore mental life as a

whole), we cannot justifiably make any substantive claims about it, let alone on its

257

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. behalf. As we already know, however, Levinas insists on pushing the matter much

farther than this logical limit, reading the proto-impression in a heterological light as the

ethical contact of the self with an absolute exteriority: indeed, it is his heterological

reading of Husserl’sUrimpression that leads Levinas to feelphenomenologically

justified in claiming that one can be touched by the alterity of the Other person in a way

that escapes the objectifying nature of perception and intentionality. When he suggests

that we can sense the “gleam” of exteriority and infinitude in the face of the Other, it is

the non-intentional experience of the proto-impression that Levinas has in mind. Prior

to cognition, prior even to the emergence of subjectivity as such, Levinas declares that

we are always already in relation to the infinitely Other, who can therefore hold us

infinitely responsible because the very constitution of our own transcendental

subjectivity is based on an originary contact with the Other’s face: thus, “the face opens

the primordial discourse whose first word is obligation” (TI 201). Beyond the reach of

reason, the face therefore also cannot be touched by the judgements of logic: “The

presentation of the face is not true, for the true refers to the non-true, its eternal

contemporary, and ineluctably meets with the smile and the silence of the skeptic. The

presentation of being in the face does not leave any logical space for its contradictory”

(TI 201). By placing the “vision” of the face far beyond the grasp of reason, Levinas

appears to believe he has guaranteed the ethics it is said to “consummate” an absolute

certainty completely immune to skeptical challenge. The first sentence of the preface to

Totality and Infinity seems to invoke the genealogical spirit of Nietzsche: “Everyone

will readily agree that it is of the highest importance to know whether we are not duped

258

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. by morality” (TI 21). Because it is neither true nor non-true, theUrimpression would,

Levinas appears to hope, preserve his ethics of the face from the ravages of a

Nietzschean critique.

Bemet, however, has tactfully rioted the obvious fact that there is no reason to

assume that Husserl’s Urimpression has anything to do with another person. It does not

even follow from Husserl’s analyses, he goes on to say, that this primary impression

must even originate from outside the self at all: it could very well be the result of a kind

of auto-affection. In Bemet’s own words:

One hardly need point out, however, that this form of alterity which is constitutive of the originary impression is not yet the alterity of another man or woman for whom I might feel responsible. There is nothing in Husserl that would permit us to conclude that it is another subject that affects me in this originary impression [Urimpression]. Even if we were to extrapolate from the letter of Husserl’s texts and admit that originary impression is indeed the experience of a hetero-affection, and not an auto affection, it would still need to be established that this initial form of temporal hetero-affection maintained an essential link with the traumatic hetero-affection by the suffering of the other person. (CCL 91)

Bemet’s understated and respectful, but I think devastating, criticism of Levinas’

elaboration of Husserl’s analysis of time-consciousness makes clear that by making

recourse to a conceptual item (the Urimpression) that, by definition, lies outside the

sphere of the conceptual, Levinas is basing his ethics upon a kind of bare X about which

we can say nothing with good reason. Even if we accept Levinas’ argument about the

sheer possibility of an experience of radical alterity, there is no reason at all for us to

accept his ethical claims on behalf of that “brute” non-conceptual Other. By feeling

259

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. compelled to place the face of the Other outside of the field of meaning in order to save

it from the supposedly violent intentionality of thought, Levinas has reduced the Other

to an empty and strictly meaningless philosophical construct (ironically, Levinas

appears to have saved the Other from the reach of theory by identifying the Other with a

theoretical posit). When understood this way, Levinas’ rhetorical recourse to the

religious concepts of epiphany and eschatoiogy appears both telling and unsurprising:

having placed the face-to-face relation outside the bounds of reason, anything

substantive he nonetheless goes on to assert about this rigorously non-rational relation is

bound to have the ring of a religious experience. What, after all, can we meaningfully

say about an intersubjective experience that cannot, by definition, be an object of

thought? Therefore, despite his repeated protestations to the contrary, Levinas’

iconoclastic approach to the face forces us to leave the field of philosophy and find

moral refuge in the dictates of religion.

The problem, however, lies not so much with Levinas’ paradoxical findings (as

logically unacceptable as they now appear to be), but in the philosophical compulsion

that originally drove him to search them out. That is, I would argue that it was by

assuming that consciousness is inherently imperialistic (enveloping others in nets of

meaning woven by the self), that he created a situation for which there was no other

solution than the iconoclastic one he eventually resorted to. For there is, according to

Levinas’ strict phenomenological understanding of intentionality, no way to relate to the

Other (even perceptually) without betraying his or her alterity. The only way to respect

the other’s face is therefore, we might say, to turn one’s own away. But was he not

260

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. driven to associate consciousness with an inability to relate ethically to the Other only

because he accepted the solipsistic Cartesian starting point bequeathed to him by the

idealist methodology of his teacher, Husserl? If the very fact of consciousness in fact

cuts us off as radically from one another as Levinas’ phenomenological understanding

of intentionality would suggest, his iconoclastic solution, however extreme it might

seem, could very well still compel theoretical consent; after all, what other way could

we then imagine the possibility of relating to other people if each of us is truly trapped

in his or her own monadic consciousness? But, as we have argued all along, this picture

of mental life is false. The problem, therefore, is not his paradoxical solution, but the

Cartesian picture of mind that underlies it.

* *

In this closing section, I would like to propose an alternative to Levinas’

iconoclastic ethics, one which also sees a deep analogy between morality and the way

we respond to the human countenance, but which follows Wittgenstein in

acknowledging—by taking at face value—the meaningfuiness of the face’s appearance.

However, I will arrive at my positive argument somewhat indirectly, for in order to

explain the motivation for, as well as to develop, this alternative conception of moral

values, I will need to consider the historical roots of some characteristic features of our

philosophical modernity: especially the interrelated conceptual distinctions between

facts and values, subjects and objects, and reason and mechanism. I will begin by

considering some of the historical developments that originally led to the modem

261

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. rupture between subject and object, a major problem for modem thought, which

Husserl’s theory of intentionality was designed to resolve.

Few ideas have had as immense an impact on the development of twentieth-

century philosophy as did Husserl’s notion of intentionality, though the directions this

idea was taken in have been quite various: just think, for example, of the very different

roles of the concept of intentionality in the writings of Heidegger, Merleau-Ponty,

Sartre, Levinas, Searle, Derrida, Marion, Taylor, Dennett and Dreyfus. Perhaps the

most important reason for the philosophical importance of the idea of intentionality is

that it appeared to offer an elegant way of healing the characteristically modem—that

is, post-Cartesian—rift between subject and object. After Descartes’ postulation of a

radical separation between “thinking” and “extended” matter, it became a difficult

conceptual problem to explain how the subject or “cogito” could ever accuratelyknow

anything about the objective world from which it was absolutely distinguished (thus the

branch of philosophy known as epistemology came to assume unprecedented

importance in this new post-Cartesian philosophical world). Prior to Descartes,

meanings were understood to inhere in a Ideologically structured nature: the logos, that

is, was still woven into the substance of the cosmos. The “material universe,” as

Charles Taylor puts it, was regarded “as a kind of medium, in which psychic contents

like heat and pain, or the supposed Forms or Species of scholastic tradition, could be

lodged or embodied or manifest themselves” (SoS 146). The perception of meaning,

according to this expressive—or, following Weber, “enchanted”—conception of nature,

was therefore unproblematic: meaning, that is, was something that might simply be

262

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. perceived. The meaning of the natural world was, we might say, self-diselosive: one

simply needed to look in order to understand. But against this pre-modem view,

Galileo and his contemporaries (Descartes included), argued that the materia! universe

should be understood mechanistically: as Galileo famously put it, the natural world is

“written in the language of mathematics,” thereby memorably expressing the new' anti-

Aristotelian view of nature that would drive the scientific revolution. The law-governed

regularities of cause and effect, rather than teleology, would from then on constitute the

kind of intelligibility proper to the natural world. However, our first-person experience

of thinking feels free, and our thoughts seem accountable only to the justificatory

structures of rationality, and therefore do not seem reducible to the causal processes of

the space of nature (our brains included), which is why Descartes quite reasonably

argued that mind cannot be identified with the body. But his substance dualism left it

mysteriously unclear how body and mind can then ever interact, such that sensations

might affect the mind and the mind move the body: as we noted earlier, his eventual

postulation of the pineal gland as the point of mind-body interaction was hardly a

satisfactory resolution of this conceptual difficulty.

Descartes, therefore, had rescued thought from the reductive net of efficient

causation, yet at the price of making the mind a radically private thing, setting the stage

for much of the drama of modem philosophy to come, A key recurring element of that

drama was the repeated investigation of how the subject could ever come to have

knowledge about anything outside itself (not simply other “subjects,” but even the most

basic elements of objective nature, such as quantity and extension). The Platonic Ideas,

263

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. once inhabiting a transcendent realm of Forms, were now stripped from that

supernatural pedestal and turned into intra-mental entities: ideas (we still use the same

word) are in our heads, as we would now say. Knowledge, then, became conceived of

as the creation of “representations” in the mind that, when true, map accurately onto

features of the external world. As Taylor describes this momentous transformation:

A representation of reality now has to be constructed. As the notion of ‘idea’ migrates from its ontic sense to apply henceforth to intra-psychic contents, to things ‘in the mind,’ so the order of ideas ceases to be somethingwe find and becomes something we build. (SoS 144; italics in original)

Philosophers would from then on frequently take it for granted that a mind (understood

as rational) can have no other source of knowledge about objective reality than by

means of constructed representations of an external world (understood as mechanistic).

As Wilfrid Sellars explained this conceptual dichotomy in his classic essay,

“Empiricism and the Philosophy of Mind,” on the one hand we have a logical space of

nature (structured by the laws of causality) and on the other, a logical space of reasons

(structured the laws of rationality).

Because of the binary nature of this basic conceptual picture, modem philosophy

(as John McDowell has forcefully argued) has compulsively oscillated between two

equally untenable pictures of the way mind relates to world: empiricism and

coherentism. According to the empiricist, knowledge can onlycount as knowledge if it

is answerable to the world which it purports to represent. McDowell explains the

importance of the idea of answerability to the empiricist as follows:

264

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Now how should we elaborate the idea that our thinking is thus answerable to the world? In addressing this question, we might restrict our attention, at least tacitly, to thinking that is answerable to the empirical world; that is, answerable to how things are in so far as how things are is empirically accessible. Even if we take it that answerability to how things are includes more than answerability to the empirical world, it nevertheless seems right to say this: since our cognitive predicament is that we confront the world by way of sensible intuition (to put it in Kantian terms), our reflection on the very idea of thought’s directedness at how things are must begin with answerability to the empirical world. And now, how can we understand the idea that our thinking is answerable to the empirical world, if not by way of the idea that our thinking is answerable to experience? How could a verdict from the empirical world—to which empirical thinking must be answerable if it is to be thinking at all—be delivered, if not by way of a verdict from (as W.V. Quine puts it) “the tribunal of experience”?111 (MW xii)

And therefore, at the far reaches of our cognitive access to the world (the place where

mind “experiences” the world, if you like), thought must ultimately touch upon, and

thereby be answerable to, some kind of pre-conceptual “Given.” It is because this brute

Given lies outside the sphere of the conceptual that our concepts can be answerable to

it, according to the empiricist. Otherwise, there would be no way of determining

whether our thoughts about the world are true or false: we could conceivably think

anything we like! Empirical experience, as Quine put it, serves as a “tribunal” that has

the power to declare a “verdict” upon our thoughts. One way of picturing this

epistemological transaction is by considering the idea of sensory impressions: according

to the empiricist view, impressions or impacts from the space of nature (outside the

111 The famous quotation from Quine is from “Two Dogmas of Empiricism,” in W.V. Quine,From a Logical Point o f View (Harvard Univ. Press, 1961), pp. 20-46, p. 41.

265

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reaim of the conceptual) are what ultimately ground and justify our beliefs (in the space

of reasons) about the external world.

But Sellars saw this is a radically incoherent picture of the nature of human

knowledge, for it confuses “happenings” in the space of nature with “justifications” in

the space of reasons. Happenings in the space of nature, that is, are governed by

efficient causation: they are written, as Galileo would have put it, in the “language of

mathematics.” The logical space of reasons, however, is structured by normative

relations such as “one thing’s being warranted, or—for the general case—correct, in the

light of another” (MW xv). As McDowell summarizes Sellars’ negative point:

“whatever the relations are that constitute the logical space of nature, they aredifferent

in kind from the normative relations that constitute the space of reasons” (MW xv). Put

differently, Sellars writes: “In characterizing an episode or a state as thatknowing of ,

we are not giving an empirical description of that episode or state; we are placing it in

the logical space of reasons, of justifying and being able to justify what one says” (cited

in MW xiv). Experience as conceived of by the empiricist, thatcannot is, serve as a

tribunal upon our conceptualizations (at best the notion of a pre-conceptual Given might

provide an exculpation for those beliefs we continue to hold anyway: “The trouble with

the Myth of the Given,” writes McDowell, “is that it offers us at best exculpations

where we wanted justification” [MW 13]). Which is just to say that nothing outside the

realm of the conceptual can justify or invalidate that which lies within the realm of the

conceptual. This argument about the incoherence of the idea of a brute Given, it should

be noted, applies just as much to the inner psychological realm as it does to the realm of

266

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. external perception. Indeed, McDowell reads Wittgenstein’s “private language

argument” precisely as an attack on the idea that non-conceptual entities of any kind can

ground our understanding of mental state terms (such as the concept of pain). As he

explains:

... the main point of the conception Wittgenstein attacks is to claim that “judgements of inner sense” are ultimately grounded on bare presences, rather than to devise a way to put the bare presences into words. If someone in the grip of the [private language] conception was convinced by an argument that language could not embrace the supposed items she insists on, she might reply that that is really just her point. If language could embrace them, that would mean they were within the conceptual sphere, and the point of acknowledging them is to acknowledge something that constrains spontaneity, which moves within that sphere, from outside. So certainly language cannot capture them; but still, it can seem necessary to insist, they are there to be pointed to as the ultimate justifications for judgements of “inner sense.” The fundamental thrust of Wittgenstein’s attack is not to eliminate the idea of a private language, which by itself would merely push the line of thought that he opposes to this point. Wittgenstein’s attack undermines even this position, which has already given up the idea of a private language, by applying the more general moral: a bare presence cannot be a ground for anything. (MW 19)

The idea of a brute given, which Sellars liked to call a mythological construct, “cannot

be a ground for anything,” whether we are thinking of introspection or external-world

perception.

On the other hand, those philosophers that McDowell labels coherentists share

with empiricists the modem belief that there is a radical distinction between the rational

mind and mechanistic world, yet they have come to appreciate the unworkability of the

idea of the “given” that underlies empiricist epistemology and so have felt compelled to

267

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. renounce the concept of empirical “experience” entirely. McDowell portrays Davidson

as an exemplary coherentist (unfairly so, I think, but that is unimportant for our

immediate concerns, since it is the basic outlines of coherentisirs in general that are at

stake), and he characterizes Davidson’s position this way:

... Davidson thinks experience can be nothing but an extra- conceptual impact on sensibility. So he concludes that experience must be outside the space of reasons. According to Davidson, experience is causally relevant to a subject’s beliefs and judgements, but it has no bearing on their status as justified or warranted. Davidson says that “nothing can count as a reason for holding a belief except another belief,” and he means in particular that experience cannot count as a reason for holding a belief. (MW 14)

McDowell agrees with Davidson’s assessment of the unintelligibility of empiricism’s

recourse to a notion of a brute given (i.e. sensory impacts) for the purposes of

epistemological justification, yet McDowell worries that Davidson’s solution produces

a picture of rationality that is equally unworkable since it is unconstrained from outside

itself: Davidson, that is, offers a picture of concepts as spinning in a frictionless void.

This latter picture, as McDowell notes, is just what gives empiricism its allure in the

first place, for it is an essential element of our understanding of thought as such that it

be answerable to how the world is, not simply accountable to other beliefs we may have

about the “world” as it is represented in our thoughts alone. Coherentism, in other

words, invites a philosophical recoil back to empiricism:

Davidson’s picture depicts our empirical thinking as engaged in with no rational constraint, but only causal influence, from outside. This just raises a worry as to whether the picture can accommodate the sort of bearing on reality that empirical content

268

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. amounts to, and that is just the kind of worry that can make an appeal to the Given seem necessary. (MW 14)

The problem with the coherentist alternative to empiricism, McDowell argues, is that it

does not adequately appreciate, and therefore does not address, the fundamental

motivation for empiricism, which is a perfectly reasonable desire that our knowledge be

answerable to a real world, and not merely reflect the structure of our own thoughts.

Coherentism, that is, finds that the idea of an epistemologically grounding Given is

unintelligible, jettisons that concept, and acts as if a nettlesome philosophical problem

had been definitively resolved. The coherentist solution, however, will inevitably leave

us dissatisfied (since it is at odds with our very understanding of what thought and

knowledge are in the first place) and philosophy will therefore continue to suffer from a

conceptual pendulum swing from one untenable position to the other.

In Mind and World, McDowell observes that as long as we hold on to the

modem belief that mind is rational and nature entirely mechanistic, we will continue to

oscillate between these two positions. It is this characteristically modem dichotomy

between reason and mechanism, he believes, that underlies the philosophical oscillation

he hopes to put to rest: that is, the idea of the “given” makes sense, and acquires its

theoretical allure, only because we are under the sway of a post-Galilean scientific

worldview. He therefore argues that we can hold on to the two equally compelling

ideas (1) that the normative structure of rationality issui generis when judged against

the law-govemed interactions of the space of nature and (2) that our thoughts are

answerable to our experience of how the world truly is, only if we let go of the radically

269

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. disenchanted view of nature as wholly mechanistic which was bequeathed to us by

Galileo and the modem scientific revolution (a theoretical position that he labels “bald

naturalism”). What the scientific image of nature—nature as governed wholly by the

laws of efficient causation—leaves out is the fact that for humans, nature includes a

conceptually saturated second nature: “Human beings acquire a second nature in part by

being initiated into conceptual capacities, whose interrelations belong in the logical

space of reasons” (MW xx). In other words, our phenomenological experience of the

world is not of a causal law-governed “environment” but ofworld a that is infused with

the concepts of whatever culture we are raised in, a maturation process of cognitive

development and socialization succinctly captured by the German notionbildung. of

Therefore, experience need not be solely equated with the causal impacts of brute

givens (as Sellars was right to argue, the idea of the Given is indeed a myth), but can be

understood as our phenomenological openness to a world thoroughly imbued with

conceptual meaning, at least as long as we have been provided with a “good enough”

upbringing and enculturation. As McDowell explains:

Once we remember second nature, we see that operations of nature can include circumstances whose descriptions place them in the logical space of reasons, sui generis though that logical space is. This makes it possible to accommodate impressions in nature without posing a threat to empiricism. From the thesis that receiving an impression is a transaction in nature, there is now no good inference to the conclusion drawn by Sellars and Davidson, that the idea of receiving an impression must be foreign to the logical space in which concepts such as that of answerability function. Conceptual capacities, whose interrelations belong in the sui generis logical space of reasons, can be operative not only in judgements—results of a subject’s actively making up her mind about something—but already in the transactions in nature that are

270

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. constituted by the world’s impacts on the receptive capacities of a suitable subject; that is, one who possesses the relevant concepts. Impressions canbe cases of its perceptually appearing—being apparent—to a subject that things are thus and so. In receiving impressions, a subject can be open to the way things manifestly are. (MW xx; italics in original)

It follows from McDowell’s conception of second nature that when we see that

something is “thus and so”, we have goodreason to believe that things are “thus and

so”: “That things are thus and so,” writes McDowell, “is the conceptual content of an

experience, but if the subject of the experience is not misled, that very same thing,that

things are thus andso, is also a perceptible fact, an aspect of the perceptible world”

(MW 26; italics in original). Or as Wittgenstein put it in theInvestigations, in a passage

that McDowell cites in support of his position: “When we say, andmean, that such-and-

such is the case, we—and our meaning—do not stop anywhere short of the fact; but we

mean: this—is—so” (PI §95, cited in MW 27; italics in original).

Suggestively, McDowell finds the closest analogue in the philosophical tradition

to his notion of second nature in Aristotle’s writings on ethics. Attempting to explain

the kind of rational-and-perceptual capacities that allow us to directlysee meaning in

the world, McDowell argues that we should take as a model for this type of rational

recognition the Aristotelian concept ofphronesis, which is usually translated as

“practical wisdom.” The person of virtuous character in Aristotle’sNicomachean

Ethics is one who, through proper upbringing, has been sensitized to the demands of

reason in the world:

For Aristotle, virtue of character in the strict sense is distinct from a merely habitual propensity to act in ways that match what

271

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. virtue would require. Virtue of character properly so called includes a specifically shaped state of the practical intellect: “practical wisdom,” in the standard English translation. This is a responsiveness to some of the demands of reason (though that is not Aristotle’s way of putting it). The picture is that ethics involves requirements of reason that are there whether we know it or not, and our eyes are opened to them by the acquisition of “practical wisdom.” So “practical wisdom” is the right sort of thing to serve as a model for the understanding, the faculty that enables us to recognize and create the kind of intelligibility that is a matter of placement in the space of reason. (MW 79)112

McDowell clearly means “practical wisdom”(phronesis) to serve as a general model for

understanding of all kinds, not strictlyethical understanding: he is trying to elaborate

the concept of “second nature” as a whole, not simply those aspects of it relevant to our

grasp of ethical values. But by taking the Aristotelian notion ofphronesis as a model

for how we can directly recognize (by analogy to visual perception) the claims of reason

in the world, McDowell’s argument suggests how we might make intelligible (and

therefore accept) a perfectly ordinary aspect of the phenomenology of moral evaluation:

the everyday fact that moral values that we perceive as features of the layout or

disposition of persons and events in the world strike usreally as there (and do not seem

112 For reference, here is a relevant passage from theNicomachean Ethics (Terence Irwin, the translator of the following passage, prefers “intelligence” to “practical wisdom” as an equivalent forphronesis ): “Intelligence[phronesis] is evidently not scientific knowledge[episteme ]; for, as we said, it concerns the last thing [i.e. the particular], since this is what is done in action. Hence it is opposed to understanding [nous]. For understanding is about the [first] terms, [those] that have no account of them; but intelligence is about the last thing, an object of perception[aistkesis], not of scientific knowledge[episteme]. This is not the preception of special objects, but the sort by which we perceive that the last among mathematical objects is a triangle; for it will stop here too. This is another species [of perception than perception of special objects]; but it is still perception more than intelligence is”(NE 1142a24-31; bracketed English text is Irwin’s). From Aristotle,Nicomachean Ethics, trans. Terence Irwin (Indianapolis: Hackett, 1985).

272

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. like merely subjective projections).11'1 As Maximilian de Gaynesford explains:

“McDowell is a realist about value, but of a sophisticated or qualified sort. And the

subtlety of his position depends on its appeal to second nature, to those rational-

conceptual capacities that subjects and agents naturally have, but because they naturally

acquire them rather than because they are naturally bom with them” (JM 163).

If we witness someone physically harming another person on the sidewalk, we

do not think to ourselves: that’s terrible, but only because I subjectively project upon

this state of affairs an arbitrarily negative moral coloring. The “wrongness” of the

violence strikes us as immediately and quite directly real: as an actual, objective feature

of the events we are witnessing, which anyone who had eyes to see could, and indeed

should, perceive. If a friend who was standing next to us observed the same violence

and expressed indifference to the situation, we would likely fault him or hernot for

perceiving a moral wrong: we would feel at that moment that we had discovered

something wrong with our friend, not simply with the situation. If at the grocery store

we witnessed a parent losing control and screaming at a young child (“You idiot!”),

causing the child to wilt and sob before our very eyes, we might sympathize with

whatever frustration caused the parent to lose control (perhaps we notice a broken jar of

jelly on the floor), but still we likely would (andshould) perceive the action as simply

wrong. Again, it is not part of the phenomenology of our experience of such a situation

to regard our moral evaluation as a subjective projection upon a state of affairs that

113 For anexcellent development of this notion by a student of McDowell’sand Martha Nussbaum’s, see Bridget Clarke, The Lens o f Character: Aristotle, Murdoch and the Idea o f Moral Perception (Ph.D. Dissertation: University of Pittsburgh, 2003).

273

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. objectively features no moral qualities. Ethical phenomenology—if it is to be

accepted—shows a world that has value as part of its very fabric.

In Ms Ethics: Inventing Right and Wrong, I.L. Mackie agrees that the

phenomenology of value shows values (not only moral but aesthetic as well) to be “part

of the fabric of the world” (Mackie 15), which is to say that they strike us as

“objective.” But because, according to Mackie, there is no way to find a place for such

values in a naturalistic ontology, thus rendering them ontologically “queer” as he liked

to put it, our evaluative phenomenology must be in “error”: values may appear real to

us, but that is only an illusion. Facts alone are real, of course. McDowell, however,

notes that Mackie’s argument achieves whatever conceptual force it has by mistakenly

presuming that for a value to be real it must be a “primary quality,” akin to solidity.

Taken this way, it is of course implausible that values are simply there to be seen like

the shape or size of the physical objects that make up the material furniture of our lives.

If we nonetheless insist that values are real, then Mackie would have a point when he

claims that our evaluative phenomenology is erroneous, however compelling that

phenomenology might seem regarded from a first-person perspective.

McDowell, however, argues that we do not need to accept Mackie’s association

of values with primary qualities (which makes them inevitably seem “queer”), and

rather suggests that we regard values as secondary qualities, akin to colors. Here is how

he defines secondary qualities, using the color “red” as an example:

A secondary quality is a property the ascription of which to an object is not adequately understood except as true, if it is true, in virtue of the object’s disposition to present a certain sort of

274

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. perceptual appearance: specifically, an appearance characterizable by using a word for the property itself to say how the object perceptually appears. Thus an object’s being red is understood as something that obtains in virtue of the object’s being such as (in certain circumstances) to look, precisely, red. (MVR 133)

Colors, like all secondary qualities, are not objective in the way that extension and

quantity are, for the definition of a secondary quality (such as “redness”) is inextricable

from our sense of how an appropriately sensitized subject wouldperceive a sample of

that color. Redness is notthere like an object’s shape or size, but it is there nonetheless.

Secondary properties, that is to say, are “dispositional” in nature. For something to be

red is for it to be such that a properly sensitized person would see it as red. But this is

not to reduce “redness” to a merely subjective property or projection, for a red object is

not red only at those times when it is being observed. It is red (and disposed to be red)

even when no one is looking at it. As McDowell explains:

Secondary-quality experience presents itself as perceptual awareness of properties genuinely possessed by the objects that confront one. And there is no general obstacle to taking that appearance at face value. An object’s being such as to look red is independent of its actually looking red to anyone on any particular occasion; so, notwithstanding the conceptual connection between being red and being experienced as red, an experience of something as red can count as a case of being presented with a property that is there anyway - there independently of the experience itself. (MVR 134)

And thus, redness is a real (though secondary) property of the object itself, and not

merely a subjective projection of a given beholder, though we cannot fully understand

the concept of redness without making reference to appropriately sensitized perceiving

275

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. subjects. As de Gaynesford sums all this up, explicitly extending the analogy from

colors to values conceived of as secondary properties:

Moral and aesthetic properties are of the sort that are not fully describable or intelligible without reference to the effect they have on subjects of experience and agents. In that sense, they are more like the colour- or taste-properties that an object might have than like those shape- or size-properties it has. All these properties are objective; but whereas shape- and size-properties are fully describable and intelligible independently of our sensitivity to them, colours, tastes and values are not. This is because colours, tastes and values are a particular kind of objective property: they are dispositions of objects to appear a certain way to suitably sensitive observers in suitable circumstances. An object’s having the colour-property ‘red’, for example, just is its being such as to look red given these circumstances. Given the same circumstances, an object’s having the aesthetic value ‘beautiful’ just is its being such as to appear beautiful, and a person’s having the ethical value ‘kind’ just is his being such as to appear kind. So the existence of such properties depends partly on the existence of beings with the capacity to observe them and pick them out. (JM 163-4)

McDowell’s development of a secondary-quality construal of values allows him to

acknowledge the dependence of values on the contingent existence of certain forms of

human life (including societies that raise those “beings with the capacity to observe

[certain values] and pick them out”) yet at the same time to accept their reality. That is

to say, like colors values are secondary qualities that are there whether or not a given

individual notices them at a given time; they are real aspects of the world that await

recognition. According to his position, as de Gaynesford explains:

Ethical and aesthetic beliefs and judgements can be true or false. Ethical and aesthetic properties really exist in the world. These values are things to which moral agents and aesthetic judges can be sensitive or insensitive. These values are really there to be discovered; they are not projected onto the world, willed into

276

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. existence or constituted by emotional reactions. They are distinct from discrete affective experiences that individuals may or may not have when confronted with them. They make a real difference to the situation or individuals that possess or lack them. (IM 167)

Given McDowell’s reliance on the analogy of value-perception to color-perception, it is

significant that there has been much dispute (and not simply in reference to his writings)

about whether or not colors themselves should be understood in a dispositional

manner.114 Perhaps, as some, like Locke, have maintained, the phenomenology of color

is itself simply illusory: Mackie, unsurprisingly, goes in for this very position. Others,

taking a physicalist line, argue that color is simply “the property of selectively reflecting

incident light at certain wavelengths,” and thus color is relational but non-sensory (cf.

JM 168). While still others take a “strong objectivist” position and “claim that objects

are coloured, and that the nature of this property is intrinsic” (JM 168).

If we could hold onto a dispositional understanding of values only if we were

able to establish the dispositional nature of colors, it would be essential that we address

and refute each of these competing ways of understanding color perception. De

Gaynesford, however, suggests that the analogy between colors and values, as

illuminating as it is in McDowell’s hands, does not need to be understood that strictly:

We can understand McDowell as saying the following: “Why shouldn’t we understand value as we understand colour - in the world, though internally related to the exercise of sensibility?” If we respond: “because that is not how colour works”, that does not undermine McDowell’s overall claim about value. Even if colour does not work that way, other things might, and they would provide an effective model for value. (JM 169)

114 S ee , e.g., de Gaynesford, pp. 168-170.

277

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I think the hypothetical response de Gaynesford suggests that we might make to the

skeptic about the dispositional nature of colors is true, but weak, for if we cannot point

to any “other things” that “might” work the way McDowell wants colors to work, then

we would have some reason to worry about the relevance and force of McDowell’s

argumentative strategy. However real values might appear to us, perhaps Maclde was

right all along. If nothing else is dispositional in the way McDowell claims values are,

then why should we believe values are that way in the first place, unless we made

argumentative recourse to the compelling force of our first-person evaluative

phenomenology? But that would hardly be a convincing move since it is that very

phenomenology that was originally at issue: after all, even Mackie agrees that values

appear real.

Whether or not colors work the way McDowell suggests (though I suspect they

do), I propose that we consider Wittgenstein’s aspect-seeing explanation of face-

perception (as I have developed it in earlier chapters, especially chapter two) as a better

model than color perception for a realist construal of the recognition of values. Indeed,

I would argue that to some extent, face perception does not simply provide a good

analogical model for understanding how moral perception might function (as construed

by McDowell and others who subscribe to a perception-based model of ethics), but that

the aspect-seeing powers that enable face perception in general are thevery cognitive

mechanisms that enable moral perception as well. Even if colors are not dispositional

in the way McDowell wants them to be, my reading of Wittgenstein has demonstrated

278

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that faces and the aspects that play across their features certainly are. There is nothing

“queer” about the fact that we can “see” the glance of the eye, for example, as

Wittgenstein caustically observed in a passage fromZettel that we looked at earlier:

“If you only shake free from your physiological prejudices, you will find nothing queer about the fact that the glance of the eye can be seen too.” For I also say that I see the look that you cast at someone else. And if someone wanted to correct me and say that I don’t reallysee it, I should take that for pure stupidity. (Z 223)

And yet, our ability tosee sight itself—like our ability to perceive value—cannot be

explained by merely broadening the reach of the senses. As Wittgenstein put it

elsewhere, aspect-seeing implies a “modified concept of sensation” (PI p. 179; my

emphasis). The “glance” is not an empirical fact like the color of an iris, or the dilation

of a pupil: “I should contradict anyone who told meI saw the glance ‘just the way’ I see

the shape and colour of the eye” (Z 223). Wittgenstein’s passing negative remark here

about the perception of eye-colour suggests the extent to which even a compelling

argument for the dispositional nature of color-perception will not yet account for how

we can see mind in a face, without which it would be difficult to imagine a satisfyingly

full account of the nature of intersubjective moral perception. That is to say, while

certainly quite illuminating, the conceptual reach of McDowell’s suggestion that values

are secondary qualities like colors may ultimately be too limited. For instance, consider

how implausible it would be to claim that we see fear the way we see the color red; and

yet could we claim to have an adequate ethics that did not account for how we

recognize (and respond to) another person’s fearful state of mind?

279

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. If we take Wittgenstein’s writing on face-perception as a model for the

perception of values, then we need not worry about the ontological status of colors. The

range of moral and aesthetic values that we recognize as part of the fabric of our world

can be regarded as just as real as the states of mind that we see flashing across the faces

of those around us (but no more real than they are either!). As we noted in chapter two,

recognizing the aspect of timidity in a face is something that one who has mastered the

concept of timidity can do without a thought, so to speak. We can have a “purely visual

concept... of a timid face”: wholly visual, yet conceptually rich. But for one who has

not mastered the concept of timidity, a timid aspect would not be directly recognizable

upon the face of another, and this “blindness” would be made evident by the fact that he

would “react to the visual impression differently”: perhaps, for example, showing

indifference (or even irritation or hostility) where another might have shown concern.

The way a person responds, or does not, to an aspect of mind in the face of another

shows whether that person can experience (that, fullyis knows the concept for) that

aspect of mind at all. I would argue that the intersubjective responsiveness that is

internal to Wittgenstein’s way of understanding the possession of mental state concepts

is already substantially ethical in nature. To fully recognize something like fear is also

to be disposed torespond with concern and care to the presence of that state of mind in

another person, though in a fashion that cannot be specificed or predicted in advance by

the application of rules. This line of thought suggests how compatible is Wittgenstein’s

understanding of physiognomy with a realist model of moral perception: for if we saw

someone responding inappropriately (whether with indifference or annoyance) to

280

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. another person whose face clearly expressed fear, we would say that there was

something at fault with the morally insensitive person’s character. Speaking in

Aristotelian terms, a person who did not recognize—and therefore did not appropriately

respond to—the fear in another person’s face could be said to suffer from an

impairment in Ms or her practical intellect. That is, blindness to the demands of moral

reasons can, and perhaps should, be understood as a form of aspect-blindness.

On the other hand, and somewhat paradoxically, the perception of the particulars

of another person’s facial expressions plays no positive role in Levinas’ ethical

phenomenology of the face. Levinas’ way of overcoming Husserl’s transcendental

solipsism shows him still to be in the grip of a fundamentally Cartesian picture of mind,

which as I have argued, explains why he would feel himself able to respect the alterity

of the other person’s face only by placing it beyond the reach of visual perception

altogether. Wittgenstein’s later writings, on the other hand, show how the expressive

meaningfulness of the face can play a positive, even essential, role in ethical life. His

theory of aspect-seeing explains not only how we can understand the expressions of a

face but also how we can perceive ethical truths, preserving the reality of moral claims

while steering clear of both ethical subjectivism and fundamentalism. Aspect-seeing,

that is, underlies not only facial perception but also practical wisdom as well (i.e.

phronesis): moral values, like the mental states in a face, are neither objective facts in

the world, nor subjective projections of the beholder, but they arereal nonetheless. I

suggest that it is because of this deep cognitive link between face perception and the

way we recognize mind and value that the human countenance has been so important to

281

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. modem aesthetic experimentation, and philosophical reflection: for face perception uses

the same powers of pattern recognition that enable us to perceive personhood and moral

truths. Indeed, it is because of the connections between face, mind and value that the

making and receiving of images of the human face can provide us ways to reflect

creatively upon the meaning of personhood in the skeptical and ethically disoriented

climate of our late modernity.

282

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Bibliography

Aristotle, Nicomachean Ethics , trans. Terence Irwin. Indianapolis: Hacked, 1985.

Ashbery, John. Selected Poems, New York: Penguin, 1986.

Attridge, Derek. JM . Coetzee and the Ethics o f Reading: Literature in the Event. Chicago: Univ. of Chicago Press, 2004.

Augustine, St. Confessions. Harmondsworth: Penguin, 1961.

Bacon, Francis and David Sylvester, ed.Interviews with Francis Bacon: 1962-1979, new and enlarged edition. London: Thames and Hudson, 1980.

Bacon, Francis.Bacon: Portraits and Self-Portraits, Ed. France Borel, intro. Milan Kundera. London: Thames and Hudson, 1996.

Baron-Cohen, Simon, and Cross, P. “Reading the eyes: Evidence for the role of perception in the development of a theory of mind,” pp. 173-186Mind in and Language 6 (1992).

Baron-Cohen, Simon. “Perceptual role-taking and protodeclarative pointing in autism,” pp. 113-127 inBritish Journal of Developmental Psychology 7 (1989).

Baron-Cohen, Simon.Mindblindness: An Essay on Autism and Theory of Mind. Cambridge, MA: MIT Press, 1995.

Bates, E., Benigni, L., Bretherton, I., Camaioni, K., and Volteira, V. “Cognition and communication from 9-13 months: Correlational findings”The in Emergence of Symbols: Cognition and Communication in Infancy, ed. E. Bates. Academic Press, 1979.

Blair, R.J.R., Frith, U., Smith, N., Abell, F., and Cipolotti, L. “Fractionation of visual memory: Agency detection and its impairment in autism,” pp. 108-118 in Neuropsychologia 40 (200).

Bloom, Paul. How Children Lean the Meaning o f Words. Cambridge, MA: MIT Press, 2000.

Bohn, Willard.Apollinaire and the Faceless Man: The Creation and Evolution of a Modem Motif. London: Associated Univ. Presses, 1991.

283

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Brook, Andrew and Don Ross, eds.,Daniel Dennett. Cambridge: Cambridge Univ. Press, 2002.

Bruce, Vicld, Ed.Face Recognition: a Special Issue of the European Journal of Cognitive Psychology 3 (1).

Brunswik, Egon. Perception and the Representative Design of Psychological Experiments, 2nd ed. Berkeley, 1956.

Butterworth, G. “The ontogeny and phytogeny of joint visual attention”Natural in Theories o f Mind, ed. A. Whiten. Oxford: Blackwell, 1991.

Clark, Maudemarie. Nietzsche on Truth and Philosophy. Cambridge: Cambridge University Press, 1991.

Clarke, Bridget. The Lens of Character: Aristotle, Murdoch and the Idea o f Moral Perception. Ph.D. Dissertation: University of Pittsburgh, 2003.

Coetzee, J.M. Waiting for the Barbarians. New York: Penguin, 1980.

Cox, Christopher Cox.Nietzsche: Naturalism and Interpretation. Berkeley: Univ. of California Press, 1999.

Critchley, Simon and Robert Bemasconi, Eds.The Cambridge Companion to Levinas. Cambridge: Cambridge Univ. Press, 2002.

Daix, Pierre. “Portraiture in Primitivism and Cubism,” pp. 254-295 in William Rubin, ed. Picasso and Portraiture. London: Thames and Hudson, 1996.

Daix, Pierre.Picasso createur. Paris: Seuil, 1987.

Davidson, Donald.Essays on Actions and Events. Oxford: Clarendon Press, 1980.

Davidson, Donald.Subjective, Intersubjective, Objective. Oxford: OUP, 2001.

de Gaynesford, Maximillian.John McDowell. Oxford: Polity, 2004.

Deleuze, Gilles. Francis Bacon : the logic o f sensation. Trans. Daniel W. Smith. Minneapolis: Univ. of Minnesota Press, 2003.

Deleuze, Gilles and Felix Guattari,A Thousand Plateaus: Capitalism and Schizophrenia, Trans. Brian Massumi. Minneapolis: Univ. of Minnesota Press, 1987.

284

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. de Man, Paul.The Rhetoric of Romanticism. New York: Columbia Univ. Press, 1984.

Dembski, William A. “Randomness by Design,” pp. 75-106 inNous 25, No. 1 (March 1991).

Dennett, Daniel.Brainstorms: Essays on Designing Minds. Cambridge, MA: M U Press, 1998.

Eliot, T.S. “The Love Song of J. Alfred Prafrock,” pp. 3-8 inThe Waste Land and Other Poems. Ed. Frank Kermode. New York: Penguin, 1998.

Elkins, James. Pictures of the Body: Pain and Metamorphosis. Stanford: Stanford Univ. Press, 1999.

Far ah, Martha J. Visual Agnosia: Disorders o f Object Recognition and What They Tell Us about Normal Vision. Cambridge, MA: MIT Press, 1990.

Farah, Martha. The Cognitive Neuroscience o f Vision. Oxford: Blackwell Publishers, 2000 .

Field, T.M., Woodson, R., Greenberg, R. and Cohen, D. (1982). “Discrimination and imitation of facial expressions by neonates,” pp. 179-81 Sciencein 281.

Frith, Uta. Autism: Explaining the Enigma, 2nd ed. Oxford: Blackwell, 2003.

Garton, Charles. Personal Aspects o f the Roman Theatre. Toronto: Hakkert, 1972.

Gombrich, E.H. “The Mask and the Face: the Perception of Physiognomic Likeness in Life and in Art,” pp. 1-46 inArt, Perception, and Reality, Ed. Maurice Mandelbaum. Baltimore: Johns Hopkins Univ. Press, 1970.

Gombrich, E.H. Art and Illusion: A Study in the Psychology o f Pictorial Presentation. London: Phaidon, 1972.

Gombrich, E.H. The Story o f Art. London: Phaidon Press, 1995.

Goren, C.C., Sarty, M., and Wu, R.W.K. (1975). “Visual following and pattern discrimination of face-like stimuli by newborn infants,” pp. 544-9 inPediatrics 56.

Graham, John. Lavater’s Essays on Physiognomy: A Study in the History o f Ideas. Berne: Peter Lang, 1979.

285

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Guttenplan, Samuel, ed.A Companion to the Philosophy o f Mind. Oxford: Blackwell, 1994.

Heidegger, Martin. Being and Time. Trans. John Macquarrie & Edward Robinson. New York: Haper, 1962.

Heider, Fritz and M. Simmeh “An experimental study of apparent behaviour.” American Journal of Psychology, 13, 1944.

Hegel, G.W.F. Hegel’s Aesthetics: Lectures on Fine Art, Vol. 1. Trans. T.M. Knox. Oxford: Clarendon Press, 1975.

Hegel, G.W.F. Phenomenology o f Spirit. Trans. A.V. Miller. Oxford: Clarendon Press, 1977.

Hietanen, J., and Perrett, D. “A role of expectation in visual and tactile processing within temporal cortex” inBrain Mechanisms of Perception and Memory: From Neuron to Behavior, ed. T. Ono et al. Oxford: OUP, 1991.

Husserl, Edmund. Cartesian Meditations: an Introduction to Phenomenology. Trans. Dorion Cairns. The Hague: M. Nijhoff, 1960.

Husserl, Edmund. Ed. Donn Welton.The Essential Husserl: Basic Writings in Transcendental Phenomenology. Bloomington: Univ. of Indiana Press, 1999.

Johnson, M. H., Dziurawiec, S., Ellis, H., Morton, J. (1991). “Newborns’ preferential tracking of face-like stimuli and its subsequent decline,” pp. 1-19 inCognition 40 (no. 1-2).

Jones, John.On Aristotle and Greek Tragedy. New York: Oxford Univ. Press, 1962.

Karmel, Pepe. Picasso and the Invention o f Cubism. New Haven: Yale Univ. Press, 2003.

Kim, Jaegwon. Mind in a Physical World: An Essay on the Mind-Body Problem and Mental Causation. Cambridge, MA: MIT Press, 1998.

Koethe, John. The Continuity o f Wittgenstein’s Thought. Ithaca: Cornell Univ. Press, 1996.

Kripke, Saul. Wittgenstein on Rules and Private Language. Cambridge, MA: Harvard Univ. Press, 1982.

286

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Lavater, Johann Caspar.Essays on Physiognomy, 3 vols. London: J. Murray, 1789-98.

Leiris, Michel. “Francis Bacon, full face and in profile,” trans. John Weightman, in Francis Bacon. New York: Rizzoli International Publications, 1987.

Leslie, Alan. “Pretence and representation: The origins of ‘theory of mind’,” pp. 412- 426 in Psychological Review 94 (1987).

Leslie, Alan. “Pretense and representation: The origins of ‘theory of mind’,” pp. 412-26 in Psychological Review (94).

Leslie, Alan. “ToMM, ToBy, and Agency: Core architecture and domain specificity” in Mapping the Mind: Domain Specificity in Cognition and Culture, eds. L. Hirschfeld and S. Gelman. Cambridge: Cambridge Univ. Press, 1994.

Levinas, Emmanuel. Totality and Infinity: An Essay on Exteriority, trans. Alphonse Lingis. Pittsburgh: Duquesne Univ. Press, 1969.

Loveland, K., and Landry, S. “Joint attention and language in autism and developmental language delay,” pp. 335-349 in Journal of Autism and Developmental Disorders 16 (1986).

Maurer, D . “Neonatal synaesthesia: Implications for the processing of speech and faces” in Developmental Neurocognition: Speech and Face Processing in the First Year of Life, ed. B. de Boysson-Bardies et al. Kluwer, 1993.

Mauss, Marcel. “A category of the human mind: the notion of person; the notion of self,” pp. 1-25, trans. W.D. Halls, in The Category o f the Person: Anthropology, Philosophy, History, eds. Michael Carrithers, Steven Collins, Steven Lukes. Cambridge: Cambridge Univ. Press, 1985.

McDowell, John. Mind and World. Cambridge, MA: Harvard Univ. Press, 1994.

McDowell, John. Mind, Value and Reality. Cambridge, MA: Harvard Univ. Press, 1998.

McGurk, H., MacDonald, J. (1976). “Hearing lips and seeing voices” Naturein (264), pp. 746-8.

Monk, Ray. Ludwig Wittgenstein: the Duty o f Genius. New York: Penguin, 1990.

Moran, Dermot. Introduction to Phenomenology. London: Routledge, 2000.

287

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Morrison, Toni. Song o f Solomon. New York: Plume, 1977.

Mundy, P., Sigman, M., Ungerer, I., and Sherman, T. “Defining the social deficits in autism: The contribution of nonverbal communication measures,” pp. 657-669 in Journal of Child Psychology and Psychiatry 27 (1986).

Napier, A. David. Masks, Transformation, and Paradox. Berkeley: Univ. of California Press, 1986.

Nichols, K., and Champness, B. “Eye gaze and the GSR,” pp. 623-626Journal in of Experimental Social Psychology 1(1971).

Nietzsche, Friedrich. The Gay Science, Trans. Walter Kaufman. New York: Random House, 1974.

Penrose, Roland.Picasso: His Life and Work. New York: Harper and Row, 1958.

Perrett, D., and Mistlin, A. “Perception of facial characteristics by monkeys” in Comparative Perception, vol. 2: Complex Signals, ed. W. Stebbins and M. Berkeley. New York: Wiley, 1990.

Phillips, W., Baron-Cohen, S., and Rutter, M. “The role of eye-contact in the detection of goals: Evidence from normal toddlers, and children with autism or mental handicap,” pp. 375-383 inDevelopment and Psychopathology 4(1992).

Pineda, Cecile. Face. Rev. ed. San Antonio: Wings Press, 2003.

Prestige, George L. God in Patristic Thought. London: W. Heinemann, 1936.

Quine, W.V.O. “Indeterminacy of translation again”The in Journal of Philosophy, vol. 84, No. 1, Jan. 1987.

Quine, W.V.O. From a Logical Point of View. Cambridge: Harvard Univ. Press, 1961.

Quine, W.V.O. The Pursuit of Truth. Cambridge, MA: Harvard Univ. Press, 1990.

Quine, W.V.O. Word and Object. Cambridge, MA: MIT Press, 1960.

Reddy, V., “Playing with other’s expectations: Teasing and mucking about in the first year” in Natural Theories of Mind, ed. A. Whiten. Oxford: Blackwell, 1991.

288

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Rhees, Rush Rhees, ed. Recollections o f Wittgenstein. Oxford: Oxford Univ. Press, 1984.

Rilke, Rainer Maria. The Notebooks ofMalte Laurids Brigge. Trans. Stephen Mitchell. New York: Random House, 1982.

Rubin, William, ed.Picasso and Portraiture. London: Thames and Hudson, 1996.

Sabbatini, Renato M.E.Phrenology: the History o f Brain Localization.

Scaife, M., and Bruner, I. “The capacity for joint visual attention in the infant,” pp. 265- 266 inNature 253 (1975).

Schaffer, H. “Early interactive development” inStudies in Mother-Infant Interaction, ed. H. Schaffer. Academic Press, 1977.

Schultz, R.T., Gauthier, L, Klin, A., et al. “Abnormal ventral temporal cortical activity during face discrimination among individuals with autism and Asperger’s syndrome,” pp. 331-340 inArchives of General Psychiatry 57 (2000).

Sellars, Wilfrid. “Empiricism and the Philosophy of Mind,” in Herbert Feigl and Michael Scriven, eds.,Minnesota Studies in the Philosophy o f Science, vol. 1, pp. 253-329. Minneapolis: University of Minnesota Press, 1956.

Shakespeare, William. Hamlet. Ed. Harold Jenkins. London: Routledge, 1982.

Shakespeare, William. Macbeth. Ed. David Bevington. New York: Bantam, 1980.

Spengler, Oswald.The Decline o f the West. Trans. Charles Francis Atkinson. New York: Knopf, 1926-8.

Stem, Daniel. The First Relationship: Infant and Mother. Cambridge, MA: Harvard Univ. Press, 1977.

Stem, Daniel. The Interpersonal World o f the Infant. New York: Basic Books, 1985.

Stewart, Susan. Poetry and the Fate of the Senses. Chicago: Univ. of Chicago Press, 2001.

Taylor, Charles. Sources of the Self: The Making o f the Modem Identity. Cambridge, MA: Harvard Univ. Press, 1989.

289

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Topffer, Rodolphe. Essai du Physionogmie in Enter the Comics, trans. E. Wiese. Lincoln: Univ. of Nebraska Press, 1965.

Vamedoe, Kirk. “Picasso’s Self-Portraits,” pp. 110-179 inPicasso and Portraiture, Ed. William Rubin. London: Thanes and Hudson, 1996.

Wada, J. “Modification of cortically induced responses in brainstem by shift of attention in monkeys,” pp. 40-42 in Science 133 (1961).

Warrington, E., and Shallice, T. “Category specific semantic impairments,” pp. 829-854 in Brain 107 (1984).

Weininger, Otto.Sex and Character, trans. of 6th German ed. London: William Heinemann, 1906.

Williams, Raymond.Keywords: A Vocabulary o f Culture and Society, rev. ed. New York: Oxford Univ. Press.

Wittgenstein, Ludwig. Culture and Value. Ed. G.H. von Wright. Trans. Peter Winch. Chicago: Univ. of Chicago Press, 1980.

Wittgenstein, Ludwig.Philosophical Investigations, 3rd ed., Trans. G.E.M. Anscombe. Oxford: Blackwell, 2001.

Wittgenstein, Ludwig.The Wittgenstein Reader, Ed. Anthony Kenny. Oxford: Blackwell, 1994.

Wittgenstein, Ludwig.Zettel, Eds. G.E.M. Anscombe and G.H. von Wright,, Trans. Anscombe. Berkeley: UC Press, 1970.

Wolff, P. “Observations on the early development of smiling” Determinantsin of Infant Behavior, vol. 2, ed. B. Foss (New York: Wiley, 1963)

Wollheim, Richard.Art and its Objects: an Introduction to Aesthetics. New York: Harper and Row, 1971.

Wollheim, Richard. Art and its Objects: with Six Supplementary Essays, 2nd Ed. Cambridge: Cambridge Univ. Press, 1980.

Wollheim, Richard. Painting as an Art. Princeton: Princeton Univ. Press, 1982.

290

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Yamane, S., Kaji, S, and Kawano, K. (1988) “What facial features activate face neurons in the inferctemporal cortex of the monkey?” inExperimental Brain Research (73), pp. 209-214.

Yin, R. K. (1969) “Looking at upside-down faces,” pp. 141-5 inJournal of Experimental Psychology (81).

Young, Andrew, Brace, Vicki. “Perceptual Categories and the Computation of ‘Grandmother’,” pp. 5-49 in Vicki Bruce, ed., Face Recognition: a Special Issue of the European Journal of Cognitive Psychology 3 (1).

Young, Andrew.Face and Mind. Oxford: Oxford Univ. Press, 1998.

291

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.